Current template: single.php

PDF

 

Fiscal Rules and Public Debt: An Emerging Market Perspective1

 

 

Pablo E. Guidotti

 

School of Government, Universidad Torcuato Di Tella

 

First draft, August 31st, 2020. Revised, November 30th, 2020

 

 

  • I wish thank Guillermo Calvo, Ricardo López Murphy, Steve Hanke, Gabriel Lopetegui, and Barry Poulson for their insightful comments.

 

Introduction

 

 

For the last four decades the world has witnessed a steady increase in global public debt. By 2020, Gopinath and Gaspar (2020) estimate that global public debt will reach its highest level ever recorded, at over 100 percent of GDP. Although public debt of emerging and developing countries has fluctuated more in relation to GDP than that of the advanced economies, today’s public debt levels are set to reach record levels in both groups of economies. Therefore, it is fair to say that global public debt levels currently pose one of the most important and urgent challenges for macroeconomic policy around the world.

 

The recent build-up of debt has reflected quite unusual economic conditions. On the one hand, the world has been hit by two unprecedentedly profound crises: the 2008 Global Financial Crisis and the ongoing Covid-19 pandemic; both have required massive fiscal policy stimulus in response. On the other hand, interest rates in mature bond markets have been reduced to record low—near zero—levels since 2008 and, in some cases, they have even gone into negative territory. Near-zero rates have created conditions favorable for a sustained growth in public debts, as governments privileged the counter-cyclical role of fiscal policy.

 

Moving forward, the task of recovering fiscal soundness will take center stage as soon as a new normal for the world economy is reached. In this context, implementation of fiscal rules at supra-national and national levels is seen as a potentially effective instrument to restore fiscal soundness. However, so far, the design of fiscal rules has not produced adequate guidance so as to contain the growth of public debts around the world.

 

 

– 1 –

 

In this paper, I discuss how existing fiscal rules relate to public debt. A main message of this paper is to warn that current benchmarks utilized in most fiscal rules in effect around the world are too lax. Especially in, but not limited to, emerging market and developing economies, the interaction of fiscal policy with the international capital market is central, and liquidity constraints that translate in sudden stops of capital flows requires adoption of public debt ceilings that are significantly lower than those adopted in most fiscal rules and in the debt sustainability guidelines used by the International Monetary Fund (IMF). Moreover, such debt ceilings relate to debt-management policies that governments can adopt—such as the choice of debt maturity—to reduce liquidity risk.

 

Adoption of tighter public debt ceilings has important implications for the resolution of fiscal problems in emerging market and developing countries and for the design of IMF programs in situations when countries face a significant loss of investors’ confidence. In particular, adequate debt sustainability analyses need to include the design of effective rules for debt restructuring.

 

The paper is organized as follows. The next section summarizes the current discussion on fiscal rules, the conceptual aspects and the empirical evidence, as well as the importance of domestic politics in determining fiscal outcomes. Th following section discusses the notion of debt sustainability for emerging market and developing countries with access to international capital markets and derives numerical safe public debt ceilings. It also discusses the implications for debt restructuring and the international financial architecture. The last section concludes.

 

 

– 2 –

 

The Debate on Fiscal Rules

 

 

Fiscal soundness has long been a central objective of macroeconomic policy. However, for a number of reasons that I will discuss in this paper, soundness of fiscal policy has often been a rather elusive concept. And especially for developing and emerging market economies the cost of unsustainable fiscal policies has been very high at least on two fronts.

 

Firstly, high deficits have often led to debt crises, as has been documented by Reinhart and Rogoff (2009). Argentina’s long default in the 2002-2016 period and the 2012 Greek debt restructuring, as well as the debt troubles faced by Argentina and Ecuador in 2020 show how difficult it is to achieve fiscal soundness in practice despite its universally recognized importance in theory.

 

Secondly, fiscal difficulties can lead to “fiscal dominance,” a situation in which central banks and monetary policy becomes intertwined with fiscal policy and dependent on it. The most obvious consequences of fiscal dominance are loss of central bank independence, loss of confidence in the currency and, eventually, high and persistent inflation. The relationship between fiscal dominance and high inflation can also be related to the presence of large public debts denominated in domestic currency, as have been studied, for instance, by Calvo (1988) and Calvo and Guidotti (1990). It is shown there how time inconsistency of monetary policy, associated with fiscal dominance, can lead to the loss of nominal anchor by the central bank, generating multiple equilibria. The presence of multiple equilibria helps us understanding why, under certain conditions, inflation can be very difficult to control, can be extremely variable, and can even result in hyperinflation. The hyperinflation experiences of

 

 

– 3 –

 

Argentina and Peru at the end of the 1980s and beginning of the 1990s stand as testimony of the dangers associated with extreme fiscal dominance.2 One of the best examples of extreme fiscal dominance in recent times is provided by the 1994 Yugoslav hyperinflation, during which the central bank financed over 95% of all government expenditures. 3

 

In the end, lack of soundness in fiscal policy results in an increase in the public debt burden. In a recent IMF blog, Gaspar and Gopinath (2020) document the historical evolution of the ratio of gross public debt to GDP in advanced and emerging market economies. Figure 1 below, taken from their article, illustrates the evolution of the public-debt-to-GDP ratio in both economic groups of countries from 1880 to 2020.

 

As can be clearly observed, in the context of the Covid-19 pandemic, the public-debt-to-GDP ratio is set to reach historical heights in both advanced and emerging market economies. Following the policy response to the Covid-19 crisis4, advanced economies are expected to reach a public-debt-to GDP ratio of over 130 percent, piling up on the already very high public debt levels reached in response to the 2008 Global Financial Crisis. In the

 

 

 

 

 

 

 

  • Argentina experienced hyperinflation in 1989 and 1990, with inflation reaching almost 4924% in 1989 and 1344% in 1990; data accessed from indec.gob.ar/indec/web/Institucional-Indec-InformacionDeArchivo-1. Peru experienced hyperinflation between 1988 and 1990, with inflation reaching 1722% in 1988, 2775% in 1989, and 7650% in 1990; data accessed from www.imf.org/external/datamapper/PCPIEPCH@WEO/OEMDC/ADVEC/WEOWORLD/PER. See Hanke and Krus (2013) for a precise characterization of each hyper-inflation episode.

 

3 See Hanke (1999).

4 According to Gaspar and Gopinath (2020), the fiscal response (both higher spending and lost revenue) to the Covid-19 pandemic has reached, on average, about 9 percentage points of GDP in the advanced economies and about 3 percentage points of GDP in emerging market economies. In addition, advanced economies deployed an additional 11 percent of GDP in loans, equity, and guarantees, compared to an additional 2 percent of GDP in these same items across emerging market economies.

 

 

– 4 –

 

case of emerging market economies, public debt is also expected to reach a historical record exceeding 63 percent of GDP.

 

Figure 1. Selected Economies: Public Debt in Relation to GDP

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

When considering the evolution of public debt to GDP in the advanced economies one cannot ignore the fact that today’s levels exceed the previous historical peak experienced of the post-World-World-Two (WWII) that had exceeded 120 percent. Moreover, as can be

 

 

– 5 –

 

observed in Figure 1, public debt in the advanced economies decreased rapidly in relation to GDP in the period following the end of WWII, until reaching a minimum in the mid-seventies.

 

The behavior of the advanced economies’ public debt after WWII mimicked the evolution of the US Federal Debt. This evidence begs the question about the role of fiscal austerity in producing a rapid and consistent—over a thirty-year period—reduction in the public-debt-to-GDP ratio after WWII and whether there are lessons to be learned about fiscal soundness. Unfortunately, the answer to this question is not very encouraging.

 

In an interesting paper, Brown (1990) analyzes precisely which were the factors that dominated the steady decline in the ratio of the US Federal Public Debt and GNP from 1945 through 1974. In 1945 the US Federal Debt amounted to 110 percent of GNP and fell to 23 percent of GDP in 1974. In his paper, Brown studies the relative role played by three factors in the reduction of the US-Federal-Debt-to-GNP ratio between 1945 and 1974 (and two sub-periods5). The three factors are: 1) real economic growth; 2) net real debt repayment (i.e., fiscal policy/austerity); and 3) net price changes (i.e., inflation).

 

Brown’s results are highly telling. When considering the period 1945-1974, fiscal policy did not contribute to reducing the US Federal Debt in relation to GNP. On the contrary, fiscal policy contributed to increasing the public debt burden by 10 percentage points—the only subperiod when fiscal policy contributed with a small 3 percent reduction was the first decade (1945-1955). The 86 percent fall in the ratio of US Federal Debt to GNP between 1945 and 1974 was due, in similar amounts, to economic growth (a 55 percent reduction) and to

 

 

  • The sub-periods analyzed by Brown (1990) are 1945-1955 and 1945-1965)

 

– 6 –

 

inflation (a 41 percent reduction). Namely, almost half of the reduction in the US debt burden was due to inflation, its effect being similarly important in all sub-periods.

 

Interestingly, the above facts are consistent with empirical findings by Calvo, Guidotti, and Leiderman (1991) on the evolution of the US Federal Debt and its maturity structure between 1946 and 1988. Especially motivated on the US Federal Debt data, Calvo and Guidotti (1992) characterized the optimal choice of debt maturity in a model where the government is subject to time inconsistency and inflation is therefore a factor that plays an important role in fiscal policy decisions. In particular, the US Federal Debt fell consistently in relation to GDP between 1946 and the mid-seventies and started to increase thereafter. Debt maturity displayed a remarkably similar behavior, it shortened consistently in the period preceding the mid-seventies and lengthened thereafter within the sample period.

 

Two main results of the Calvo and Guidotti (1992) model are particularly important. Firstly, in a time-inconsistency context, government displays “debt aversion”; namely, the government’s inability to pre-commit its policies (in particular, inflation) raises the cost of debt and, hence, introduces a bias in the optimal policy toward debt reduction. Secondly, management of the maturity structure of the public debt is central to the implementation of the optimal policy. The model predicts a positive association between the public-debt-to-GDP ratio and debt maturity. Calvo, Guidotti and Leiderman (1991) find that the stylized Calvo and Guidotti (1992) model conforms with US data.

 

The discussion so far simply suggests that the debt reduction episode occurred after WWII may not be useful to deal with the current surge in public debts around the world. In

 

 

– 7 –

 

particular, one of the main developments in macroeconomic policy that took place from the early 1980s and especially with the advent of financial globalization in the 1990s has been the defeat of inflation and the concentration of central banks on achieving price stability as its main—and often only—monetary policy goal. Such evolution in monetary policy was supported by theoretical developments; for instance, Barro and Gordon (1983) laid the basis for an extensive literature on the importance of “central bank independence”.6

 

With inflation increasingly out of the policy picture, containment of rising public debt ratios was left to just two factors: economic growth and fiscal soundness. Thus, in the 1990s the concept of fiscal rules—and the adoption of numerical constraints on fiscal policy— started to take center stage in the general discussion about fiscal soundness and debt sustainability.

 

Fiscal rules emerge as instruments to design a fiscal policy that is consistent with low inflation and monetary policy independence and, thus, prevents fiscal dominance. But very importantly, fiscal rules are seen as instruments to force politics to accept the principle of fiscal discipline. Perhaps the most famous example of fiscal rules established to discipline politics is, in the European context, the Treaty of Maastricht signed in 1992, later complemented by the adoption of the Stability and Growth Pact in 1998. These fiscal institutions responded primarily to the perception that fiscal indiscipline had been rampant in Europe during the 1970s and 1980s, prompting high and growing public debts as well as often pro-cyclical fiscal policies in a number of European Union (EU)’s members. Therefore, the

 

 

  • Cukierman (1992) contains an excellent discussion about the theoretical underpinnings of central bank independence.

 

– 8 –

 

Treaty of Maastricht and the Stability and Growth Pact were seen as instruments to guide convergence of fiscal policies within the EU, especially because fiscal indiscipline was deemed as inconsistent with the European Monetary Union and, in particular, the creation of the European Central Bank (ECB) and the Euro as the block’s single-currency.7 By being supra-national, the European fiscal rules were expected to enjoy strong political support and enforceability at the national level.

 

In emerging markets, as can be observed in Figure 1, public debt ratios have increased since the 1980s to reach a historical record under the effects of the current Covid-19 crisis. However, the limited access to the international capital market as well as the occurrence of various sovereign debt crises played central roles in containing the growth of public-debt-to-GDP ratios in the emerging market economies relative to those observed in the advanced economies.

 

The resolution of the 1980s external debt crisis with the implementation of the Brady Plan at the beginning of the 1990s jumpstarted a process of significant integration of emerging market economies to the international capital market—known as “globalization”. The Brady Plan was crucial in fostering financial globalization not only because it put an end to decade-long series of defaults across Asia, Africa, and Latin America; its implementation implied the securitization of large stocks of non-performing loans owed by governments to a

 

 

 

 

 

 

 

 

  • See Buti and Giudice (2004).

 

– 9 –

 

set of large international banks. Securitization implied the transformation of illiquid bank loans into Brady bonds that could be widely traded in capital markets.8

 

The Brady Plan allowed risk to be unloaded from the balance sheet of international banks and transferred to a variety of institutional and retail investors in the international capital market. Issuance of Brady bonds by a large number of emerging market economies allowed investors to price and trade risk, which started to be measured by indices such as the Emerging Market Bond Index computed by J.P. Morgan. Once sovereign risk was priced and traded in the capital market, it became easier for governments and corporates to issue additional debt.

 

As governments recovered access to the international capital market and policy frameworks were strengthened by the reform agenda promoted by the new Washington Consensus, central banks in emerging market economies were able to pursue the, until then, elusive task of winning the battle against persistent high inflation. And, indeed, by the first half of the 1990s, inflation had quickly fallen to single digits in most of the developing and emerging-market world.

 

With inflation out of the picture and with a renewed reliance of governments on external foreign-currency debt financing, fiscal indiscipline would rapidly translate into unsustainable public debt levels. And excessive growth in public debt also fueled by an exuberant optimism of international capital markets with the advent of globalization. But

 

  • The Brady Plan resulted in a massive securitization of previously illiquid bank assets. This eventually led to the emerging market crises of the second half of the 1990s and beginning of the 2000s. A decade later, a similar process occurred in the advanced economies, when the securitization of sub-prime mortgages in the US became a central cause of the 2008-2009 Global Financial Crisis.

 

 

-10-

 

market sentiment is known to shift suddenly and without major changes in fundamentals. As a reassessment of risk by investors took hold in the second half of the 1990s, a series of crises rocked emerging markets and some (foreign and domestic) public debts ended up being defaulted and restructured—the most important cases being Mexico, Russia and Argentina.9

Motivated by the financial turmoil in emerging markets, the adoption of fiscal institutions in Europe and the successful launching of the Euro, the international discussion on fiscal rules picked up steam in the 1990s. Following Schaechter el al. (2012), fiscal rules are defined as “a long-lasting constraint on fiscal policy through numerical limits on budgetary

 

aggregates.”10 So-called first generation rules started to be adopted in the 1990s, initially in Europe with the Treaty of Maastricht in 1992 and the Stability and Growth Pact in 1998. Adoption of fiscal rules in developing countries started to become popular mostly in the late 1990s in response to the sequence of financial crises that rocked emerging markets.

 

In terms of design, fiscal rules take various forms that differ from country to country. Some fiscal rules set limits on budget deficits, others on government expenditure or revenues, and some set targets in relation to the public debt. Several countries adopted legislation—often referred to as Fiscal Responsibility Laws—to enforce fiscal rules and make them credible across different political administrations. In some countries, the fiscal framework also comprises the setting up of independent fiscal institutions, known as Fiscal Councils.

 

 

  • Russia and Argentina defaulted in 1998 and 2002, respectively. Mexico did not suffer a full-fledged default because it was rescued by the US Treasury in 1995 to resolve a short-term debt (Tesobonos) crisis.

10 Fiscal rules were firstly defined by Kopits and Symansky (1998) as “a permanent constraint on fiscal policy, typically defined in terms of overall fiscal performance.”

 

 

-11-

 

As discussed by Eyraud et al. (2018), the growing popularity of fiscal rules in both the advanced and the developing and emerging economies since the late 1990s was also accompanied by the perception that rules were becoming too complex, lacked enough flexibility in the wake of shocks, and were difficult to enforce. This led to the birth of second-generation rules in the aftermath of the 2008 Global Financial Crisis with the emphasis placed at making fiscal rules more flexible and more enforceable albeit, sometimes, at the expense of simplicity. Currently, over 90 countries around the globe have in place fiscal rules of different types.11

 

Before briefly reviewing the existing three-decades-long empirical evidence on the effectiveness of fiscal rules, it is useful to discuss the conceptual justification behind their adoption, as well as the associated trade-offs. The principal conceptual justification for the adoption is the political-economy argument that democratically elected governments often tend to display a “deficit bias” either because the electoral process tends to make governments more myopic as they focus on maximizing their chances of political success, or because governments are unable to commit policies of future administrations. 12

Moreover, once excessive deficits are in place the political process may make fiscal adjustment more difficult or delayed in time, as argued by the “war of attrition” model of Alesina and Drazen (1991). In turn, political delays in implementing needed fiscal adjustment may spill over to monetary and exchange-rate instability. Guidotti and Végh (1999) rationalize the interaction of fiscal policy with exchange-rate stability. Combining Alesina and

 

  • See Schaechter el al. (2012) and Eyraud et al. (2018).
  • See, for instance, the discussion in Persson and Tabellini (1990).

 

 

-12-

 

Drazen (1991)’s “war of attrition” model with a balance-of-payments crisis model, Guidotti and Végh (1999) characterize “credibility” of an exchange-rate stabilization plan as the probability that a needed fiscal consolidation would be agreed by the political process before the occurrence of a balance-of-payments crisis. Guidotti and Végh (1999) show that credibility displays an inverted-U shape; while credibility increases in the initial stages of stabilization, it will fall if fiscal consolidation is delayed long enough, making the failure of stabilization in the midst of a foreign exchange crisis increasingly likely.

 

Interaction between fiscal policy and monetary stability is also a major element justifying the adoption of fiscal rules in the context of supra-national institutional frameworks. The most obvious case is provided by Europe, where the adoption of fiscal rules reflected the need to coordinate national fiscal policies in a way that, in the aggregate, they were consistent with the EMU, the single currency and an independent ECB focused on maintaining price stability. Providing incentives to avoid negative spillovers within the EMU was particularly important in view of the existing heterogeneity in fiscal policy stances and public debt levels across Europe previous to the signing of the Treaty of Maastricht.13

 

Perhaps the most difficult task in designing and implementing fiscal rules relates to the issue of flexibility. The notion that fiscal policy has to be designed to be able to respond flexibly to shocks that affect the economy is central to conventional theory. Based on Barro’s (1979) seminal contribution, conventional theory characterizes optimal fiscal policy as one that is characterized by two fundamental principles. Firstly, optimal fiscal policy should seek

 

 

 

  • See Buti and Giudice (2004).

 

-13-

 

to raise a government revenue through a tax structure that is stable over time—i.e., Barro’s (1979) “tax smoothing” principle—while meeting an intertemporal government solvency constraint. Stability of the tax structure seeks to smooth tax distortions over time and, consequently, is optimal by, for instance, facilitating consumption smoothing by individuals.

 

Secondly, while meeting intertemporal solvency, fiscal policy should respond counter-cyclically to shocks affecting the economy. On the one hand, government revenues vary directly with economic activity and, hence, are affected by the economic cycle. In particular, there may be negative shocks affecting economic activity that depress government revenues. In the conventional framework, optimal fiscal policy should let the deficit widen in response to temporary negative shocks while debt should be reduced in the wake of positive shocks.

 

On the other hand, in addition to their effect on fiscal revenues, shocks that affect the economy affect government expenditures too. As in the case of revenues, government spending is affected automatically by economic activity through programs such as unemployment insurance and social assistance. Such programs expand or contract endogenously in response to economic activity and contribute to the counter-cyclicality of fiscal policy.14

 

Non-cyclical shocks that affect the economy, or tail events such as the 2008 Global Financial Crisis or the Covid-19 pandemic, require ad-hoc government spending programs in response. To the extent that such programs are designed to be temporary in nature their implementation will be expected to expand the deficit during the duration of the shock and

 

  • Factors that affect government revenues and established expenditure programs through changes in economic activity are called “automatic stabilizers”.

 

 

-14-

 

revert when the economy enters the recovery phase, adding to the counter-cyclicality of fiscal policy.

 

While counter-cyclicality is widely recognized as a desirable feature of fiscal policy, it is much less clear how this concept should translate into specific numerical deficit ceilings and the appropriate setting of escape clauses. By characterizing optimal fiscal policy as designed to be consistent with an intertemporal government budget constraint—thus, assuming solvency—the conventional tax-smoothing model provides little guidance about which levels of public debt in relation to GDP are safe and which are not. In the conventional framework, policy flexibility is large and, for instance, any path that ensures a stable public-debt-to-GDP ratio over time meets the requirement of government solvency. In turn, different stable public-debt-to-GDP ratios relate to combinations of deficits and economic growth rates.

 

The above-mentioned lack of guidance of the conventional optimal fiscal policy framework is reflected in the international experience about fiscal rules. For instance, rather than being anchored in theory, the numerical targets set in the Treaty of Maastricht were the result of political negotiations to accommodate a quite heterogenous set of initial fiscal positions while inducing a needed fiscal consolidation in the Euro area.15 The Treaty of Maastricht set a deficit ceiling of 3 percent of GDP with a number of escape clauses. Escape clauses allowed the budget deficit to exceed the 3 percent ceiling in circumstances deemed “exceptional and temporary”—such as a severe economic downturn with a fall in GDP of at

 

 

  • As reported by Buti and Giudice (2004), the Treaty of Maastricht induced a 3.5 percentage points fiscal consolidation between 1993 and 1997 in the Euro area, starting from an historical initial budget deficit of 5.5 percent of GDP. By 1997, all member countries except Greece had a budget deficit below the ceiling of 3 percent of GDP.

 

 

-15-

 

least 2 percent. In addition, the Treaty of Maastricht set a 60 percent reference value for the ratio of public debt to GDP for member states. For countries where the debt-to-GDP exceeded the reference value, the stance of fiscal policy would be set so as to approach the 60 percent target over time.

 

The numerical targets set in the Treaty of Maastricht continue to have significant influence around the world. According to the IMF (2018), 92 countries had fiscal rules in place by 2015, of which more than 60 percent were emerging market and developing economies.16 The large majority of countries adopting a public debt ceiling or target set it between 60 and 70 percent of GDP, mostly reflecting supra-national fiscal rules. The EMU and the East Caribbean Currency Union have a debt target of 60 percent of GDP, while the Central African Economic and Monetary Community and the West African Economic and Monetary Union adopted a debt ceiling of 70 percent of GDP. Excluding countries with supra-national rules, most other economies currently adopt a threshold of 60 percent of GDP. No country has a public debt threshold below 40 percent of GDP.

 

The numerical ceilings of the Treaty of Maastricht also continue to be dominant when referring to the budget deficit in relation to GDP. The vast majority of countries with fiscal rules adopted a 3 percent ceiling on the overall fiscal deficit in relation to GDP.17

In sum, absent a clear theoretical guidance about public debt targets, international experience has remained anchored on public debt thresholds initially proposed by the supra-

 

 

 

  • By 2017, according to the IMF Fiscal Rules Database, the total number of advanced, emerging-market, and developing economies employing fiscal rules had increased to 96.
  • See IMF (2018).

 

 

-16-

 

national rules adopted by the EMU. In the next section I will argue that, especially in the case of emerging market and developing economies, debt thresholds in the 60-70 percent of GDP range are too high. The argument will be based on the relation between a level of public debt and what it implies in terms of access to the international capital market.

 

Lack of theoretical guidance about what adequate debt-ceiling levels is also present in the IMF’s recommendations about how to design fiscal rules. In particular, when discussing the design of a fiscal rule, the IMF (2018) distinguishes between countries where a maximum debt limit is known from those were the such limit is unknown. The maximum debt limit is defined as a public debt level in relation to GDP “beyond which a debt distress episode will occur with heightened probability (for instance, default, restructuring, or large increases in sovereign spreads).”

 

The notion of an unknown debt ceiling is particularly telling. According to the IMF (2018), this case “is most suitable for advanced economies with unconstrained market access, where considerable uncertainty might exist about how much debt can be sustained.” Namely, advanced economies are assumed to conform with the conventional optimal fiscal policy framework discussed above.

 

The IMF (2013) guidelines for public debt sustainability propose public debt benchmarks of 70 percent of GDP for emerging market economies and 85 percent of GDP for advanced economies and that “these benchmarks should not be construed as levels beyond which debt distress is likely or inevitable, but rather as an indication that risks increase with the level of indebtedness.” With these benchmarks in mind, assessment of debt sustainability is

 

 

-17-

 

computed by using a stochastic simulation model—the IMF’s Debt Sustainability Analysis (DSA)—so as to ensure that the country does not exceed the maximum limit over the medium term with high probability. 18 Therefore, the maximum debt benchmark or limit plays a central role anchoring the fiscal policy framework. 19 If such limit is set too high, the tension between fiscal policy flexibility and debt sustainability is likely to be resolved in favor of higher public debt levels. As mentioned above, flexibility required by recent global crises has so far resulted in record public debt. In emerging market economies, the recent cases of Argentina and Ecuador are examples of IMF programs where, soon after their signing, events of debt default or restructuring occurred even though performance criteria were largely met.

 

Before turning to further discussing debt sustainability in the context of emerging markets, it is useful to briefly review the conclusions of recent empirical studies about the effectiveness of second-generation fiscal rules. Given the large variety of national and supra-national fiscal rules in place, setting numerical targets on government expenditure, revenue, budget balance and debt as well as escape clauses, it is quite difficult to assess their effectiveness in practice.20

 

 

 

 

 

 

 

  • In addition to the debt benchmark, the IMF (2013) guidelines and the IMF’s DSA analysis take into account a number of other relevant benchmarks such as bond spreads, external financing requirements, the change in short-term public debt in percent of total debt and the share of public debt held by non-residents.

 

  • In the midst of the Covid-19 pandemic, the IMF (2020b) has initiated a review towards reforming its Public Debt Limits Policy (DLP) in Fund supported-programs. The reform proposals aim at improving measurement, transparency and disclosure of public debt information, as well as increasing flexibility in the design of debt conditionality so as to “strike the right balance between providing space for public investment to support inclusive growth and maintaining debt sustainability”.

 

  • Schaechter el al. (2012) contains a detailed discussion of the different types of fiscal rules adopted by countries around the world.

 

 

-18-

 

A recent report prepared by the IMF Staff—see Eyraud et al. (2018)—summarizes the most recent empirical findings regarding the question of fiscal rules’ effectiveness. The first observation is that, considering outcomes of fiscal deficit and general government debt observed between 2000 and 2015, on average fiscal rules appear to influence fiscal policy in the right direction, albeit by a limited amount. When comparing the set of countries with fiscal rules with the set of countries that does not have fiscal rules, the improvement in fiscal balances and public debt in the former group, on average, is smaller than 0.2 and 5 percent of GDP, respectively.

 

The second important observation relates to the effectiveness of fiscal rules in reducing the “deficit bias”. In this respect, more detailed studies that recognize the vast heterogeneity across countries suggest the existence of a “magnet effect”; namely, rules tend to affect low and high deficit countries in opposite direction. For instance, with a numerical limit for the budget deficit set at 3 percent of GDP, empirical evidence suggests that such limit tends to work not as a maximum level but as an attractor. Countries with deficits below 3 percent of GDP tend to increase their deficits while countries with deficits exceeding 3 percent tend to reduce them.

 

Additional relevant empirical results are found in the context of the EMU. In particular, compliance with fiscal rules and the excessive fiscal deficits impact financial perceptions. Empirical evidence summarized by Eyraud et al. (2018) suggest that excessive deficits may result in higher bond spreads between 50 and 150 basis points. Also referring to the EMU, Gaspar (2020) shows that compliance has been poor, especially since the 2008 Global

 

 

-19-

 

Financial Crisis. Since 2009, between 70 and 90 percent of EMU members did not comply with one or two fiscal rules.

 

In sum, empirical evidence suggest that fiscal rules have had some impact towards improving fiscal performance, but the effect has not been large, and compliance has not been strong. Moreover, in the absence of a strong theoretical guidance about “safe” public debt levels, the challenges imposed by recent global crises have translated into growing public debts around the world. While advanced economies appear to be in a position to handle large public debts in relation to GDP, emerging market economies remain vulnerable to international capital market volatility. Hence, growing public debts in emerging markets require re-thinking about tighter public debt ceilings or benchmarks, an issue I will discuss in more detail in the next section.

 

Finally, and especially in emerging and developing economies, it has to be recognized that the empirical assessment on the effectiveness of fiscal rules depends critically on the domestic political environment and dynamics. In fact, adoption of fiscal rules is by definition a political decision that usually involves the Executive Branch as well as Congress. Moreover, in order to be successful, fiscal rules have to be maintained and enforced through different administrations, often of opposite political views.

 

Rather than discussing this issue at an abstract level, it is useful to briefly compare two contrasting experiences in Latin America: Argentina and Peru.21 As mentioned above,

 

 

  • When discussing fiscal rules in the Latin American context, the most obvious example that comes to mind is Chile—see, for instance, Frankel (2011) and Marshall (2003). Chile pioneered fiscal rules in the region adopting a structural fiscal balance target—accounting not only for the economic cycle but also for the evolution of the

 

-20-

 

Argentina and Peru are countries that share similar initial conditions marked by profound monetary and fiscal instability at the end of the 1980s. Between 1989 and 1990, both Argentina and Peru were immersed in hyperinflation and were part of the group of developing countries that had defaulted on their public debt during the 1980s. Moreover, both countries displayed among the highest levels of financial dollarization in the region.22

 

But soon after resolving their debt problems by entering the Brady Plan—Argentina in 1993 and Peru in 1995—both economies entered a period of profound macroeconomic and structural reforms and economic liberalization under the presidencies of Menem in Argentina and Fujimori in Peru. Fiscal consolidation and new budget methodologies to consolidate budget management, accounting, and treasury operations of the government, ambitious privatization in several sectors, trade openness, and monetary stabilization helped turning around both economies after decades of instability. Following the example of Chile, both Argentina in 1993 and Peru in 1992 implemented a wide-ranging reform of their social security systems, moving to an individual-capitalization system anchored in the development of private pension funds.23 As in other countries in the region, the social security reform was one of the most important drivers of the development of stable and deep domestic capital markets, although its implementation placed significant short-term pressure of fiscal accounts.24

 

 

 

price of copper—that has been very successful over time. I focus here on the comparison between Argentina and Peru because they share fundamental similarities before diverging in terms of fiscal outcomes.

  • See Guidotti and Rodriguez (1992).

 

  • While Argentina moved from a pay-as-you-go regime to a private individual-capitalization system, in Peru’s reform envisaged a coexistence of both the public and the private pension system.

 

  • Guidotti (2006) analyzes in detail the impact of Argentina’s 1993 pension reform on its fiscal dynamics.

 

 

-21-

 

In 1999, Congress in both Argentina and Peru passed a Fiscal Responsibility Law, aiming at strengthening fiscal policy and adopt a medium-term framework conducive to fiscal soundness. By the year 2000, both Argentina and Peru displayed similar levels of public debt in relation to GDP at just under 45 percent of GDP, slightly below the debt ratios observed in South America (at 50.7 percent of GDP) and emerging market and developing economies (at 47.7 percent of GDP).25 Also the fiscal deficit was similar—slightly above 2 percent of GDP— and largely in line with levels observed in South America (at 2.4 percent of GDP) and in emerging market and developing economies (at 1.6 percent of GDP).

 

The common ground shared by Argentina and Peru ended there, as domestic politics began to interfere heavily with Argentina’s economic policy decisions. In the year 2000, under the presidency of Fernando De La Rua, the Argentine government modified the Fiscal Responsibility Law by raising the deficit ceiling, and in 2001 simply violated the new limit, showing lack of political consensus in relation of fiscal rules. At the beginning of 2002, in the midst of a deep political crisis that generated a succession of five presidents in a matter of just a few days, Argentina entered a full-fledged economic crisis. The government devalued the Peso, declared default on the public debt and forced an arbitrary pesification of contracts in the economy.26

 

Since the 2001-2002 crisis, Argentina never re-established fiscal rules. On the contrary, in 2006, the government nationalized the pension system effectively destroying the domestic

 

  • Data from the IMF’s Data Mapper.
  • Pesification refers to the compulsory conversion of US-dollar contracts into domestic currency. In the banking system, pesification was “asymmetric,” as bank assets were converted into pesos at a lower exchange rate than that applied to the conversion of deposits. For an in-depth analysis of the 2001-2002 Argentine crisis, see Guidotti and Nicolini (2016).

 

 

-22-

 

capital market. As it will discussed in the next section, the default occurred 2002 ended only in 2016. After just four years, in 2020, Argentina defaulted again—for the ninth time in its history—and reached a new public debt restructuring in September, the second in just seventeen years. As Argentina allowed the return of money printing to finance fiscal deficits high inflation returned. As a result, liability dollarization remained high despite the 2002 pesification experiment. In 2019, 70 percent of Argentina’s public debt was denominated in foreign currency, mostly the US dollar. A similar level of dollarization remains in place after the 2020 public debt restructuring. 27

 

In contrast, Peru stayed the course of fiscal responsibility and passed a new Fiscal Responsibility and Transparency Law in 2003, maintaining a deficit ceiling of 1 percent of GDP and adopting rules to limit the growth on nonfinancial government expenditures.28 Maintaining fiscal responsibility throughout different political administrations was reflected in Peru’s economy. Indeed, Peru’s fiscal and economic performance in recent years has been impressive. In the 18-year period running between 2002 and 2019 average economic growth in Peru amounted to 5.3 percent per year, compared to a yearly average growth rate of South America of 2.8 percent per year. Peru’s gross public debt in relation to GDP declined to a level of 27 percent of GDP in 2019, one of the lowest in the region.

 

In terms of fiscal policy, Peru was able to sustain a significant improvement in its general government’s overall fiscal balance as a percent of GDP. From a deficit position of 2.1% of

 

  • As discussed in Guidotti (2020), Argentina’s latest debt restructuring may not solve the country’s debt woes. The structure of the 2020 restructuring has not reduced the significant debt burden; it has only postponed payments into the next administration. Moreover, fiscal dominance remains significant, as the 2021 budget contemplates a monetary financing of 60% of the budget deficit.

 

  • See Rossini, Quispe and Loyola (2011) and Liendo (2015) for an analysis of fiscal rules in Peru.

 

 

-23-

 

GDP in 2000 and 2001, Peru was able to reach a surplus of 2% of GDP in 2006, and 3.3% and 2.7% of GDP in 2007 and 2008. Over the period 2002-2019, Peru maintained on average a deficit of 0.2% of GDP per year compared to an average deficit of 3.6% a year in South America and an average deficit of 2% per year among emerging market and developing economies. 29

 

Peru’s reputation as a fiscally responsible sovereign also translated in a solid performance in terms of monetary policy. Since recovering from hyper-inflation Peru’s central bank was able to a low and stable rate of inflation. Over the period 2002-2019, Peru’s average inflation stood at 2.8% per year, compared to an average inflation rate of 5.8% per year among emerging market and developing economies. Price stability and fiscal soundness allowed Peru to develop the domestic bond market, and liability dollarization was significantly reduced over time to a 37 percent of the total public debt in 2019. 30

 

Peru’s success story in terms of its economic management has been reflected in the evolution in Peru’s credit ratings over the past two decades. Each of the principal credit ratings agencies determined that Peru’s sovereign bonds warranted investment grade status in 2008.31 Since then, Peru has maintained its investment-grade category and its credit ratings further improved over time.32

 

 

 

  • Data from the IMF Data Mapper.
  • See Peru’s Ministry of Finance, Estrategia de Gestion Integral de Activos y Pasivos 2019-2022, 2019.
  • In April and July 2008, respectively, Fitch and S&P raised Peru’s credit rating to BBB-, in the investment-grade category. Moody’s followed suit in December 2009 when it assigned Peru the investment-grade rating of Baa3.

 

  • Peru’s public debt is currently rated A3 by Moody’s and BBB+ by Standard & Poor’s and Fitch. On April 16th, 2020, in the midst of the Covid-19 crisis, Peru issued debt with 11 years maturity at an annual interest rate of 2.8 percent.

 

 

-24-

 

In sum, the contrasting experiences of Argentina and Peru show the decisive importance of domestic politics in determining the success of fiscal rules, independently of the merits of their design. Political consensus maintained throughout several administration of differing views is, in the end, perhaps the most important ingredient that separates success from failure.

 

Public Debt and Fiscal Rules in Emerging Markets

 

 

A central lesson learned from emerging-market economies’ integration to international capital markets is that defining fiscal sustainability is a complex task. It involves—as suggested by the IMF’s DSA framework—looking at a variety of factors that eventually affect the trajectory of public debt and have significant consequences on investors’ judgment about an emerging market country’s future capacity to repay its liabilities. Such factors—much less important when considering advanced economies—include the presence of liquidity constraints, of liability dollarization, and of institutional weaknesses and informality that weakens the capacity to raise government revenues.

 

In a context of potential sudden stops of capital flows, it becomes particularly difficult to define public debt sustainability, as solvency and liquidity considerations become intertwined. Moreover, sudden stops of capital flows often force fiscal policy to behave pro-cyclically, in contradiction with the recommendations of the conventional optimal fiscal policy framework discussed in the previous section.

 

Guidotti (2007) presented a simple framework to analyze how the interaction of liquidity and solvency considerations impinges on public debt sustainability in emerging markets. In

 

-25-

 

particular, when governments are not fully credible in the eyes of investors—as reflected in risk premia and credit ratings—the notion that fiscal policy can be designed under the assumption that governments meet an intertemporal budget constraint may lose practical relevance.

 

In particular, both governments—through their annual budget laws—and investors tend to evaluate a country’s capacity to access the capital market over relatively short horizons, most typically on a yearly basis.33 In this context, as pointed out in Guidotti (2007), the concept of gross public-sector borrowing requirement—measuring the size of the yearly financial program—becomes a highly relevant variable. Over this variable is where both investors and the government’s agency in charge of public debt management will focus when designing the borrowing program. In this respect, gross borrowing needs—defined as the sum of the overall budget deficit plus the rollover of maturing debt—are more relevant that the more commonly used ratio of public debt to GDP. However, as I will show next, a limit imposed by the capital market on the size of a government’s financing program has strong implications for public debt sustainability.

 

A simple model illustrates how public debt sustainability is affected by liquidity considerations. The evolution over time of the ratio of public debt to GDP, b, is given by:

 

 

(1)
b º d nb,

 

 

 

 

 

  • Greenspan (1999) and Guidotti (2000 and 2003) pointed out the necessity of linking the concept of fiscal sustainability to the development of adequate liquidity management strategies in emerging market economies.

 

 

-26-

 

where d denotes the overall budget deficit as a proportion to GDP, n denotes a (constant) growth rate of GDP, and a dot over a variable denotes its rate of change over time. Similarly, the yearly gross borrowing requirement (net of any pre-funding) as a proportion to GDP, x, can be approximated in the following way:

 

x º d + b (2)
m
   

 

where m denotes the average maturity of the public debt.34 Equation (2) simply states that the yearly borrowing requirement is the sum of the budget deficit plus debt amortizations.

 

Given the above definitions, we are interested in obtaining an enhanced measure of sustainability where fiscal policy—in addition to being consistent with inter-temporal solvency—also satisfies a liquidity constraint of not exceeding a maximum yearly borrowing requirement, x0 . Thus, in this enhanced definition a sustainable fiscal policy satisfies

 

equation (2), where ≤ !, and the following relationship between long-run growth and the budget deficit implying that the ratio of public debt to GDP, b, is held constant over time:35

 

d = nb. (3)

 

 

As a result, the following relationship between sustainable a public-debt-to-GDP ratio, average debt maturity and the maximum gross borrowing requirement in relation to GDP obtains:

 

 

  • It is assumed for simplicity’s sake that amortizations are uniformly distributed over time.
  • The overall budget deficit, d, is related to the primary balance, s, by = − , where r denotes the rate of interest on the public debt. Together with equation (3), it yields the commonly used requirement for achieving a stable debt-to-GDP ratio; i.e., = ( −   ) .

 

 

-27-

 

 

  m (4)
b £ bmax  º x0   .  
1 + nm  

 

 

For the sake of completeness, assuming equation (4) is binding, namely x = x0 , debt

 

dynamics are described by equations (1) and (2), which combined yield:

 

 

ab, (5)
b = x0

 

where a º m1 + n.

 

 

Equation (4) is central to the analysis. The first observation is that, unlike the conventional fiscal policy framework that applies mostly to the advanced economies, the presence of liquidity constraints imposed by the capital market yields a definite ceiling for the ratio of public debt to GDP. 36

 

The second observation is that, for given x0 and n, equation (4) provides a fiscal sustainability criterion that relates the public debt ceiling to the maturity structure of the public debt. In particular, in order to ensure that a country has an adequate liquidity position vis-à-vis the capital markets, as measured by the yearly borrowing requirement, there is an inverse relationship between average debt maturity on the one hand, and the sustainable budget deficit and long-run debt-to-GDP ratio on the other. Equations (3) and (4) show that

 

 

 

  • The maximum borrowing requirement referred in the present context, although exogenous to the government, is not an arbitrarily fixed number. As discussed in Guidotti (2007) it depends on a number of characteristics, such as the economic growth rate (n), the extent of liability dollarization, the government’s credit history and a number of additional consideration such as those taken into account in the IMF (2018)’s debt sustainability guidelines.

 

  • 28 –

 

the shorter the maturity of the public debt is, the smaller is the maximum allowable deficit and long-run debt to GDP ratio.

 

The third observation of the above simple framework is that, unlike what would be desirable in the conventional framework, a tightening in the liquidity constraint—i.e., a sudden stop in financing as measured by a reduction in !—will make the fiscal response to the shock pro-cyclical, as discussed in detail by Guidotti (2007). Moreover, the pro-cyclicality of fiscal policy in a sudden stop situation may be reduced by a lengthening of debt maturity, m. However, a significant change in m over a short time period can only be achieved through voluntary debt exchange operations or through a debt restructuring process. The main difference between these two alternatives lies in the terms of the debt reprofiling; voluntary exchanges may result in higher interest costs and may eventually turn out to be counterproductive while a debt restructuring may result in a reduction in the debt burden, albeit at a potentially significant reputational cost.

 

In order to bring this analysis closer to practice, it is interesting to explore what values for the public debt ceiling are reasonably suggested by equation (4). In order to undertake this task, it is useful to first consider the recent evolution of capital flows to emerging markets. As reported in the most recent IMF’s Global Financial Stability Report (2020), size of debt inflows to emerging markets have been decreasing since the 2008 Global Financial Crisis. In particular, the cumulative debt inflows to emerging markets during inflow-episodes has been decreasing systematically from level above USD 400 billion before the 2013 Fed’s Taper Tantrum, to cumulative inflows ranging between USD 165 billion and USD 60 billion since

 

  1. Although it is hard to separate between supply and demand factors, this trend points -29-

 

towards a potentially significant reduction in the participation of emerging market debt markets in investors’ international portfolios.37 Such conclusion is consistent with the recent increase in sovereign spreads observed in a number of emerging market economies with significant access to the international capital market. For instance, between January 2018 and April 2020, sovereign risk spreads increased from 210 to 383 basis points in Brazil, from 270 to 393 basis points in Mexico, from 140 to 296 basis points in Russia, from 219 to 732 basis points in South Africa, and from 267 to 833 basis points in Turkey.38

 

Liquidity considerations appear to have played a role in shaping debt management in emerging markets, especially in the context of near-zero international interest rates developed since the 2008 Global Financial Crisis. The average maturity of public debt among emerging market economies has lengthened since 2009, increasing from an average of 5.2 years over the period 1995-2008 to 6.9 years in 2019.39 However, averages disguise significant heterogeneity across countries. Average debt maturity ranges from lows of 1 and 3.5 years in Pakistan, Egypt, Hungary, respectively, to highs between 10 and 13 years in Chile, Peru, and South Africa.40 At the same time, governments in several emerging market economies sought to develop domestic bond markets. Examples of such strategy in Latin America notably include Brazil, Chile and Peru.

 

 

 

 

 

 

37See Lane and Milesi-Ferretti (2017) for a in-depth analysis of recent trends in international financial integration.

  • See IMF Fiscal Monitor (2020).

 

  • See IMF Fiscal Monitor (2020).
  • In some cases, such as Argentina, a long public debt average maturity reflects events of debt restructuring rather than the result of an active debt management strategy.

 

 

-30-

 

These considerations suggest that “safe” public debt ceilings can be computed using ranges of debt maturity and borrowing requirements exemplified in Table 1. In particular, I consider ranges of average debt maturity that goes from 3 to 9 years and a maximum size of the yearly financing program—as measured by the government’s borrowing requirement—in a range from 5 to 9 percent of GDP. In this respect, it is important to take these values as levels that can be sustained over time under various market conditions, and not values that can be reached occasionally during periods of high investors’ optimism. In its recent debt sustainability analysis for Argentina, for instance, the IMF (2020a) reaches the conclusion that—based on the experience of a number of emerging market economies that suffered debt restructuring events—government gross financing needs should not exceed 5 percent of GDP on average over the medium term (i.e., after 2024) and 6 percent of GDP on any given year.41

 

Table 1: Implied Public Debt Ceilings

 

 

    Maximum Borrowing Requirement  
      (in % of GDP)  
    5% 7% 9%
  3 13.6% 19.0% 24.4%
Average Debt 5 21.3% 29.8% 38.3%
Maturity (in yrs.) 7 28.1% 39.4% 50.6%
  9 34.2% 47.9% 61.6%
Average   24.3% 34.0% 43.7%

Note: Computations are performed using a 3.5 percent growth rate (n).

 

 

 

 

 

 

 

  • See IMF (2020a).

 

-31-

 

The main lesson that emerges from this numerical exercise is that safe public-debt-to-GDP are much lower than those currently embedded in fiscal rules across the world (and emerging markets in particular). For instance, a 7-percent-of-GDP borrowing requirement and a 3 to 9-year average debt maturity yield, on average, a public debt ceiling in the order of 35 percent of GDP. Such level is half the size of the benchmark used by the IMF in its debt sustainability guidelines discussed in the previous section.

 

Figure 2 shows current levels of public debt in relation to GDP in Latin America and in selected countries in the region in 2019. As can be observed, in 2019 the region as a whole displayed a public debt level that amply exceed the implied safe ceiling, previous to a further deterioration occurring in 2020 as a result of the Covid-19 pandemic. Latin America’s notable exceptions are Chile and Peru with public debt levels in relation to GDP below 30 percent.

 

Figure 2. Public Debt in Latin America, 2019

 

(in percent of GDP)

 

100                                                                                      
90                                                                                      
80                                                                                      
                                                                                     
70                                                                                     Public
60                                                                                    
                                                                                    Debt/GDP
                                                                                   
50                                                                                    
                                                                                     
40                                                                                     35% safe
30                                                                                     ceiling
20                                                                                      
                                                                                     
                                                                                     
10                                                                                      
                                                                                     
0                                                                                      
  America   Chile Ecuador     Uruguay
  Argentina Brazil   Mexico Peru  
Latin   Colombia        
             

 

Source: IMF Fiscal Monitor, 2019.

 

 

 

 

 

 

-32-

 

Interestingly, a 35-percent-of-GDP safe public debt ceiling for emerging market economies is consistent with Reinhart, Rogoff, and Savastano (2003)’s conclusions from their empirical analysis on what they have named “debt intolerance”. Empirical “safe” debt ceilings suggested by their study lie below the 35 percent of GDP threshold. Reinhart, Rogoff, and Savastano’s study is highly illustrative about how significant the challenge is for Latin America’s governments in the future, as history suggests that very few countries have been successful in implementing significant debt reductions without resorting to defaults or other forms of involuntary restructuring processes. 42

 

The main conclusion from the above discussion is both straightforward and important. Once liquidity constraints deriving from countries’ interaction with international capital markets are considered, sustainable public debt levels are likely to be significantly lower than those recommended by the IMF (2018) guidelines, as well as those embedded in most current fiscal rules in place.43 Hence, the introduction of liquidity constraints in the conventional tax-smoothing model delivers a stricter notion of debt sustainability than when those considerations are not accounted for. Especially in the case of emerging market economies, it provides a realistic argument in favor of the adoption of low debt ceilings when designing fiscal rules. Moreover, it is useful to define such ceilings in relation to variables such as the

 

 

 

 

  • The quality of fiscal austerity measures is very important. Alesina, Favero and Giavazzi (2019) provide extensive evidence comparing fiscal adjustment programs based on expenditure reductions with those anchored on tax increases.

 

  • Most public debt sustainability analyses rely on the concept of gross public debt in relation to GDP. However, the precise definition of public debt to be used may be important when assessing risks in specific cases, as gross public debt includes intra-government debts as well as debt owed to multilaterals. At the same time, in some cases it is important to add central-bank debt—usually not included in the gross public debt definition—to the relevant aggregate used to assess debt sustainability.

 

 

-33-

 

average maturity (or duration) of the public debt so as to induce governments to have the right incentives to reduce liquidity risks.

 

The above simple model suggests that governments in emerging market economies should strive towards lengthening the maturity structure of their public debts. However, although this objective has been greatly facilitated by the long period of low international interest rates initiated since the 2008 Global Financial Crisis, experience shows that governments often resort to issuing debt denominated in foreign currency when they attempt to develop long-term debt markets. A tradeoff then arises, as the sustainable development of long-term domestic currency debt markets is often made prohibitive by weak credibility of monetary policy.

 

Having argued in favor of safe public debt ceilings in the 35 percent of GDP range, a fourth important observation follows from the analytical framework presented above. What happens when a country that enjoys high investors’ credibility faces a sudden loss of confidence?

 

As I have discussed above, economies that are credible vis-á-vis the capital market may operate (at least for some time) under the conventional optimal fiscal policy framework and, therefore, consider intertemporal solvency as the relevant criterion to be met, without consideration to liquidity constraints. This is the framework that currently appears as the most relevant to characterize fiscal policy in the advanced economies. And in such a world we have seen that public debt may increase without a clear bound; by 2020, advanced

 

 

 

 

 

-34-

 

economies’ public debt has reached an average of 122.4 percent of GDP and 137.7 percent of GDP among the G-7 countries.44

 

Therefore, when a country transitions from a high credibility scenario to a liquidity constrained scenario, the sudden drop in what is perceived as a sustainable debt-to-GDP ratio may be very large—from, say, from a 100 percent to GDP to a 35 percent of GDP ratio—and , effectively, impossible to implement through fiscal austerity alone. In such cases, debt restructuring becomes the only possible alternative to regain debt sustainability.

 

Although there are numerous examples of sudden stops across emerging markets, even some advanced economies have been subject to the situation described above. The best example is provided by the 2012 European sovereign debt crisis. In such crisis, a large and unsustainable build-up of public debt had forced Greece into a debt restructuring under the auspices of the ECB, the European Community and the IMF. The Greek restructuring was the first event of this type in the Euro Area and, in particular, had to be undertaken without any role for monetary policy as what was referred as the Grexit option was not legally available within the EMU and, eventually, was perceived as an even costlier alternative by the Greek authorities. 45

 

In any event, the Greek debt restructuring was deemed by the European authorities to be a “unique and non-repeatable” case, in a clear effort to avoid financial contagion to other EMU economies. However, by the end of 2011 and in 2012, contagion could no longer be contained and sovereign spreads in Ireland, Portugal, Italy and Spain surged to levels

 

  • See IMF Fiscal Monitor (2020).
  • Grexit referred to the possibility of Greece abandoning the Euro Area.

 

 

-35-

 

comparable to those observed in emerging markets during the financial crises of the 1990s. The difference this time was that sovereign spreads spiked in a context of record very low international interest rates.

 

The European policy response was decisive and successful in containing a further escalation of financial contagion. It implied major changes in the context of the EMU. Firstly, the ECB modified its policy and was allowed to intervene in bond markets so as to prevent financial conditions to differ significantly across member states. 46 Secondly, the European Stability Mechanism (ESM) was created with the ability to also intervene in bond markets to assist economies under contagion, and impose adjustment programs in exchange for its intervention, with the participation of the IMF.

 

Anchored on the analytical framework discussed above, my conclusion from the 2012 European sovereign debt crisis is that, faced with a loss of investors’ confidence, it would have been impossible to restore stability in Europe without further debt restructurings in a number of countries. And the occurrence of additional credit events would have probably expanded contagion even more. Instead, Europe had the possibility, and created the instruments to avoid new debt restructurings within the EMU by choosing to socialize the public debt problem and restore financial stability.

 

Of course, the type of policy response that was possible in the EMU is not available in emerging markets. If unsustainable public debts exceed significantly safe ceilings, the

 

 

 

  • Corsetti and Dedola (2016) discuss how the type of backstop for government debt provided by the ECB in the midst of the 2012 European sovereign debt crisis may be successful in preventing sovereign defaults while being consistent with the monetary policy objective of maintaining price stability.

 

-36-

 

likelihood of defaults and restructurings will increase significantly. Contrary to what occurred in the 1990s, when the world was surprised by financial contagion and responded treating the crisis as a liquidity problem, the notion of debt sustainability is better understood today, as is the need for debt restructuring when public debt accumulation becomes excessive.

 

Interestingly, Argentina’s 2018 exceptional-access stand-by program with the IMF is a reminder of the fact that, once investors’ confidence and market access is lost, liquidity assistance only postpones the final day of reckoning. Despite implementing a massive fiscal adjustment—mostly through government expenditure cuts—and nearly restoring primary fiscal balance in 2019, Argentina was headed for default even before the outbreak of the Covid-19 pandemic.

 

The above discussion suggests that, as the introduction of safer—and tighter—debt ceilings is adopted in debt sustainability analyses, new IMF programs may increasingly rely on a combination of liquidity-assistance-cum-debt-restructuring. Hence, adoption of standards that ensure efficiency and equitable and fair treatment in sovereign debt restructuring processes becomes a necessary component to be taken into account in the design of fiscal rules. Such standards will affect the conduct of fiscal policy and may provide strong incentives toward fiscal soundness and restraint in excessive debt buildups.

 

Since the 1990s, the international community has made significant progress in improving the resolution of international sovereign debt crises. In the absence of agreement on a formal sovereign debt resolution formal framework in place—such as Krueger’s (2002) Sovereign Debt Restructuring Mechanism (SDRM) proposal—the international community

 

 

-37-

 

focused on introducing innovations to international bond debt contracts so as to facilitate eventual restructurings and limit the role of holdout creditors in debt exchanges. Agreement was reached on the adoption of enhanced Collective Action Clauses (CACs) in international sovereign bond contracts allowing a majority of bondholders to bind all bondholders as to the financial terms of a restructuring or debt exchange offer.47 Although legal discussions are still taking place as regards to the practical implementation of enhanced CACs, there is significant optimism about their effectiveness following the debt restructurings of Argentina and Ecuador.48

 

Adoption of enhanced CACs in foreign-law sovereign bond contracts is anchored on a set of best practices over which the international community has reached consensus. The set of best practices (or, the Principles) that guide sovereign debt restructurings include that: 1) the debtor should attempt to maintain payments to creditors while a workout is being negotiated;

 

  • creditors and the debtor should negotiate in good faith; 3) creditors of equal standing should be treated equally; and 4) in order to ensure transparency and sustainability, the negotiation process should involve the IMF.49

Notwithstanding the adoption of enhanced CACs in emerging markets and the consensus reached about best practices, the issue of sovereign debt crisis resolution is complex and, when credit events take place, they become highly contentious and politically charged, and have profound consequences on the debtor country’s economy and welfare.

 

 

  • See, for instance, IMF (2014), Sobel (2016), IMF (2019)
  • See discussion in Buchheit and Gulati (2020).
  • Variations of these principles have been proposed in different international fora; see, for instance, IIF (2013) and IMF (2014).

 

-38-

 

On the one hand, issuing debt governed by the law of a foreign jurisdiction, especially in a major financial center, has important benefits for the issuer. By issuing debt governed by the law of the major financial centers, sovereigns seek to tap a larger demand for their bonds and, most importantly, “borrow” the credibility enjoyed by the legal systems of those jurisdictions. The effect of this is clear; the issuing State seeks to receive better terms on its debt and to enjoy a larger volume of financing than would otherwise be the case if it issued debt under its own domestic law.50

On the other hand, issuing debt governed by foreign law in principle reduces significantly a government’s discretion for restructuring its foreign obligations—is precisely this limit on government discretion what makes these assets more credible and desirable in the eyes of investors. But limiting government discretion comes at a cost; i.e., sovereigns (as opposed to corporations) do not enjoy the protection typically given by bankruptcy law. And here is where CACs enter into action, as a mechanism to facilitate sovereign debt restructurings by limiting the role of holdouts in debt exchanges.

 

However, the emphasis placed by the international community on the effectiveness of the enhanced CACs is implicitly based on the assumption that sovereigns will abide by the best practices mentioned above. Unfortunately, this may be not always the case.

 

Argentina’s debt saga between the 2002 default and its resolution in 2016 shows that it cannot be taken for granted that sovereign debtors will abide by best international practices. As its 2005 unilateral debt exchange offers received limited support by bondholders, Argentina

 

  • There is ample evidence that emerging market bonds governed by foreign law trade at tighter spreads than similar bonds issued under domestic law.

 

 

-39-

 

decided to openly defy foreign courts and international tribunals.51 In March 2014, the Argentine Supreme Court decided that a ruling obtained in a foreign court could be brought to enforcement in Argentina.52

 

The possibility that sovereign may choose to defy foreign courts to which they had originally submitted when issuing bonds, may weaken the credibility offered by international financial centers—such as New York—and, hence, generate negative externalities on other emerging-market issuers. Moreover, by ignoring rulings of foreign courts, rogue debtors may have negative effects on the functioning of the international capital market. This discussion suggests that, short of adopting a formal sovereign debt resolution framework, the future international reform agenda should promote the adoption of a specific debt restructuring protocol by which sovereigns agree to submit to the ruling of foreign courts and, for this purpose, renounce to the immunity currently enjoyed by their international reserves (present and future).53

 

Finally, signing of the proposed debt-restructuring protocol should be made a condition for issuing bonds under foreign jurisdiction. Countries that opt not to be signatories of the protocol would only be able to issue debt under domestic jurisdiction.

 

 

 

 

 

 

 

 

 

 

  • In 2010, Argentina reopened the 2005 debt exchange. However, the new offer still left a significant number of holdout creditors.
  • Argentine Supreme Court Ruling C. 462 XLVII, “Claren Corporation c/ E.N.—arts. 517/518 CPCC exequatur s/ varios,” March 2014.

 

  • Other aspects of this proposal are discussed in Guidotti and Hamilton (2015). In particular, in order to foster enforcement of the protocol, the BIS should be required to provide no immunity protection with respect to international reserves deposited by non-complying countries.

 

 

-40-

 

Concluding Remarks

 

 

This paper has discussed the interaction fiscal rules, the public debt, and the capital market. It has been emphasized that the presence of liquidity constraints that emerge from the interaction with the capital market has important implications for the adequate design of the fiscal policy framework and, hence, of fiscal rules.

 

In particular, it has been shown that, especially in emerging and developing economies, fiscal policy should be designed to include safe public debt ceilings that are significantly lower that those imbedded in current debt sustainability analyses. Moreover, at current high levels of public debt, it may be increasingly difficult to avoid debt restructuring as part of the multilaterals’ effort to restore financial stability in situations where governments lose the confidence of investors. In this context, further innovations may be needed in the international financial architecture to make sovereign-debt-crisis resolution more efficient.

 

 

 

 

 

-41-

 

References

 

 

Alesina A. and A. Drazen, “Why Are Stabilizations Delayed?” American Economic Review, (1991): 1170-88.

 

Alesina A., C. Favero, and F. Giavazzi, Austerity: When it Works and When it Doesn’t, Princeton University Press, 2019

 

Barro R., “On the Determination of the Public Debt,” Journal of Political Economy, (1979):

 

940-971.

 

Barro R. y D. Gordon, “A Positive Theory of Monetary Policy in a Natural Rate Model,” Journal of Political Economy, (1983): 589-610.

 

Brown C., “Episodes in the Public Debt History of the United States,” in R. Dornbusch y M.

 

Draghi, Public Debt Management: Theory and History, MIT Press, 1990.

 

Buchheit L. and M. Gulati, “The Argentine Collective Action Controversy,” Working Paper, July 2020.

 

Buti M. and G. Giudice, “EMU’s Fiscal Rules: What Can and Cannot Be Exported,” in Kopits G. (ed.) Rules-Based Fiscal Policy in Emerging Markets: Background, Analysis, and Prospects, Palgrave-Macmillan, 2004.

 

Calvo G., “Servicing the Public Debt: The Role of Expectations,” American Economic Review,

 

(Sept. 1988): 647-661.

 

Calvo G. and P.E. Guidotti, “Indexation and Maturity of Government Bonds: A Simple Model,” in R. Dornbusch y M. Draghi, Public Debt Management: Theory and History, MIT Press, 1990.

 

Calvo G. y P.E. Guidotti, “Optimal Maturity of Nominal Government Debt: An Infinite-Horizon

 

Model,” International Economic Review, (Nov. 1992): 895-919.

 

Calvo G., Guidotti, P.E. and L. Leiderman, “Optimal Maturity of Nominal Government Debt:

 

The First Tests,” Economic Letters, Vol. 33, (April 1991): 415-421.

 

 

 

 

-42-

 

Corsetti, G. and L. Dedola, “The “Mystery of the Printing Press” Monetary Policy and Self-fulfilling Debt Crises,” Working Paper, July 2016.

 

Cukierman A., Central Bank Strategy, Credibility, and Independence: Theory and Evidence, MIT Press, 1992.

 

Eyraud L., Debrun X., Hodge A., Lledó V. and C. Pattillo, Second-Generation Fiscal Rules: Balancing Simplicity, Flexibility, and Enforceability, IMF Staff Discussion Note SDN/18/04, April 2018.

 

Frankel, J., “A Solution to Fiscal Procyclicality: The Structural Budget Institutions Pioneered by Chile,” NBER Working Paper 16945, April 2011.

 

Gaspar V., “Future of Fiscal Rules in the Euro Area,” Workshop on Fiscal Rules in Europe:

 

Design and Enforcement, DG ECFIN, Brussels: Jan. 2020.

 

Gopinath G. and V. Gaspar, “Fiscal Policies for a Transformed World,” IMF Blog, July 10th, 2020.

 

Greenspan, A., “Currency Reserves and Debt,” Remarks Before the World Bank Conference on Recent Trends in Reserves Management, Washington, D.C., April 1999.

 

Guidotti, P.E., “On Debt Management and Collective Action Clauses,” in Kenen, P. y A. Swoboda (eds.), Reforming the International Monetary and Financial System, Washington: International Monetary Fund, 2000.

 

Guidotti, P. E., “Toward a Liquidity Management Strategy for Emerging Market Economies,” in Gonzalez,J., Corbo, V. Krueger, A. and Tornell, A. (eds) Latin American Macroeconomic Reforms: The Second Stage, The University of Chicago Press, 2003.

 

Guidotti P.E., “Argentina’s Fiscal Policy in the 1990s: A Tale of Skeletons and Sudden Stops,” published in Challenges to Fiscal Adjustment in Latin America: The Cases of Argentina, Brazil, Chile and Mexico, The Organization for Economic Co-Operation and Development, Chapter 3, 2006: pp. 69-92.

 

 

 

 

 

-43-

 

Guidotti P.E., “Global Finance, Macroeconomic Performance, and Policy Response in Latin America: Lessons from the 1990s,” Journal of Applied Economics, Vol. X, No. 2, (Nov. 2007), pp. 279-308.

 

Guidotti P.E., “Argentina on the Brink, Again,” OMFIF, October 7, 2020. Available at:

 

https://www.omfif.org/2020/10/argentina-on-the-brink-again/.

 

Guidotti P.E. and J. Hamilton, “Sovereign Debt Restructuring and the Global Financial Architecture,” in Uzan M. (ed.), Bretton Woods: The Next 70 Years, New York: Reinventing Bretton Woods Committee, 2015, pp. 135-144.

 

Guidotti P.E. and J.P. Nicolini, “The Argentine Banking Crises of 1995 and 2001: An Exploration into the Role of Macro-Prudential Regulations,” Prepared for the Riksbank Macroprudential Conference, Stockholm: June 2016.

 

Guidotti P.E. and C.A. Rodriguez, “Dollarization in Latin America: Gresham’s Law in Reverse?,” IMF Staff Papers, Vol. 39 No. 3, September 1992, pp. 518-544.

 

Guidotti P.E. and C.A. Végh, “Losing Credibility: The Stabilization Blues,” International

 

Economic Review, (Feb. 1999): 23-51.

 

Hanke, S., “Yugoslavia Destroyed Its Own Economy,” The Wall Street Journal, April 28, 1999.

 

Hanke S. and N. Krus, “World Hyperinflations,” in Parker R. and R. Whaples (eds.) The

 

Handbook of Major Events in Economic History, London, UK: Routledge, 2013.

 

Institute of International Finance, Principles for Stable Capital Flows and Fair Debt Restructuring, Report on implementation by the Principles Consultative Group, Washington, DC: October 2013.

 

International Monetary Fund, Staff Guidance Note for Public Debt Sustainability Analysis in Market-Access Countries, May 2013.

 

International Monetary Fund, Strengthening the Contractual Framework to Address Collective Action Problems in Sovereign Debt Restructuring, Washington: October 2014.

 

 

 

 

-44-

 

International Monetary Fund, Fiscal Policy: How to Calibrate Fiscal Rules, A Primer, How To Note 8, March 2018.

 

International Monetary Fund, Fourth Progress Report on Inclusion of Enhanced Contractual Provisions in International Sovereign Bond Contracts, March 2019.

 

International Monetary Fund, Global Financial Stability Report, April 2020.

 

International Monetary Fund, Fiscal Monitor, April 2020.

 

International Monetary Fund, Technical Assistance Report—Staff Technical Note on Public Debt Sustainability, Country Report 20/83, March 2020a.

 

International Monetary Fund, Reform of the Policy on Public Debt Limits in IMF-Supported Programs, IMF Policy Paper, November 2020b.

 

Kopits G. and S. Symanski, Fiscal Rules, IMF Occasional Paper 162, 1998.

 

Krueger, A., A New Approach to Sovereign Debt Restructuring, International Monetary Fund, Washington, DC: April 2002.

 

Lane P. and G.M. Milesi-Ferretti, “International Financial Integration in the Aftermath of the Global Financial Crisis,” IMF Working Paper WP/17/115, May 2017.

 

Liendo, C., “Evolution of Fiscal Rules in Peru,” in Werner A. and A. Santos (eds.), Peru: Staying

 

the Course of Economic Success, Washington: International Monetary Fund, 2015.

 

Marshall, J., “Fiscal Rule and Central Bank Issues in Chile,” in Fiscal Issues and Central Banking in Emerging Economies, BIS Papers No. 20, October 2003.

 

Persson T. and G. Tabellini, Macroeconomic Policy, Credibility and Politics, Harwood Academic Publishers, 1990.

 

Reinhart C. and K. Rogoff, This Time Is Different: Eight Centuries of Financial Folly, Princeton University Press, 2009.

 

Reinhart C., Rogoff K. and M. Savastano, “Debt Intolerance,” Brookings Papers on Economic Activity, Vol. 1 (Spring 2003): 1-74.

 

 

-45-

 

Rossini R., Quispe Z. and J. Loyola, “Fiscal Policy Considerations in the Design of Monetary Policy in Peru,” BIS Papers No. 67, 2011.

 

Schaechter A., Kinda T., Budina N. and A. Weber, “Fiscal Rules in Response to the Crisis— Toward the “Next-Generation” Rules. A New Dataset,” IMF Working Paper WP/12/187, July 2012.

 

Sobel M., “Strengthening Collective Action Clauses: Catalysing Change—The Back Story,” Capital Markets Law Journal, Vol. 11, No 1, 2016

-46-

Fiscal Rules and Public Debt

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.