Tag Archives: Sovereign debt

“Financial Policy,” CEPR, 2018

CEPR Discussion Paper 12755, February 2018. PDF. (Personal copy.)

This paper reviews theoretical results on financial policy. We use basic accounting identities to illustrate relations between gross assets and liabilities, net debt positions and the appropriation of (primary) budget surplus funds. We then discuss Ramsey policies, answering the question how a committed government may use financial instruments to pursue its objectives. Finally, we discuss additional roles for financial policy that arise as a consequence of political frictions, in particular lack of commitment.

“Sovereign Bond Prices, Haircuts, and Maturity,” NBER, 2017

NBER Working Paper 23864, September 2017, with Tamon Asonuma and Romain Ranciere. PDF. (Local copy.)

Rejecting a common assumption in the sovereign debt literature, we document that creditor losses (“haircuts”) during sovereign restructuring episodes are asymmetric across debt instruments. We code a comprehensive dataset on instrument-specific haircuts for 28 debt restructurings with private creditors in 1999–2015 and find that haircuts on shorter-term debt are larger than those on debt of longer maturity. In a standard asset pricing model, we show that increasing short-run default risk in the run-up to a restructuring episode can explain the stylized fact. The data confirms the predicted relation between perceived default risk, bond prices, and haircuts by maturity.

“Sovereign Bond Prices, Haircuts, and Maturity,” CEPR, 2017

CEPR Discussion Paper 12252, August 2017, with Tamon Asonuma and Romain Ranciere. PDF. (ungated IMF WP.)

Rejecting a common assumption in the sovereign debt literature, we document that creditor losses (“haircuts”) during sovereign restructuring episodes are asymmetric across debt instruments. We code a comprehensive dataset on instrument-specific haircuts for 28 debt restructurings with private creditors in 1999–2015 and find that haircuts on shorter-term debt are larger than those on debt of longer maturity. In a standard asset pricing model, we show that increasing short-run default risk in the run-up to a restructuring episode can explain the stylized fact. The data confirms the predicted relation between perceived default risk, bond prices, and haircuts by maturity.

“Sovereign Bond Prices, Haircuts, and Maturity,” IMF, 2017

IMF Working Paper 17/119, May 2017, with Tamon Asonuma and Romain Ranciere. PDF.

Rejecting a common assumption in the sovereign debt literature, we document that creditor losses (“haircuts”) during sovereign restructuring episodes are asymmetric across debt instruments. We code a comprehensive dataset on instrument-specific haircuts for 28 debt restructurings with private creditors in 1999–2015 and find that haircuts on shorter-term debt are larger than those on debt of longer maturity. In a standard asset pricing model, we show that increasing short-run default risk in the run-up to a restructuring episode can explain the stylized fact. The data confirms the predicted relation between perceived default risk, bond prices, and haircuts by maturity.

Puerto Rico’s Debt Restructuring

On Econofact, Daniel Bergstresser provides background information on Puerto Rico’s debt crisis. From his text:

  • Unlike U.S. municipalities, a U.S. territory cannot resort to Chapter 9 of the Bankruptcy act.
  • The island’s economy benefited from corporate tax exemptions (until 2006) and from tax exemptions on interest paid by municipal bonds issued by Puerto Rico and its agencies (“triple tax exemption”).
  • Total bond indebtedness (face value) amounts to over $70 billion, about 70 percent of the island’s GDP. The island owes an additional $50 billion in unfunded pension obligations to its state employees and retirees. Different government-sponsored entities issued the debt, apparently representing different claims on the Commonwealth’s revenue streams.
  • Puerto Rican issuers were downgraded from investment grade status in 2014. In March 2015, governor Padilla announced that the island’s debt was unpayable.
  • Resolution has been delayed by disagreement about the borrowers’ capacity to repay.
  • Puerto Rico defaulted on its general obligation debt in June of 2016 and President Obama signed the “Puerto Rico Oversight, Management, and Economic Stability Act” (PROMESA) law. This created an oversight board with the authority to oversee the island’s budget and facilitate restructuring talks. The law also created a bankruptcy-like “Title III” mechanism. The oversight board placed a moratorium on debt collection by the island’s creditors until May 1, 2017. On May 3 the island entered the debt restructuring process. Chief Justice John Roberts has assigned the case to U.S. District Judge Laura Taylor Swain, and the first hearing in that case is scheduled to occur on May 17.

Re-Denomination Risk in France and Italy

On the FT Alphaville blog, Mark Weidemaier and Mitu Gulati argue that re-denomination risk in the Euro zone is most prominent in France and Italy. Bonds with CACs trade at higher prices.

Most French and Italian [but not Greek] debt is governed by local law. … the governments could pass legislation redenominating their bonds from euros to francs or lira.

… [But] some French and Italian bonds — bonds issued after January 1, 2013, with maturities over a year — have Collective Action Clauses (CACs). … Importantly, these CACs require a super-majority of investors (in principal amount) to approve any changes to the currency of the bond.

… But it’s also possible a local law bond is no different than a local law bond with a CAC. After all, both are ultimately subject to the whims of the local legislature, and the courts may side with them.

The markets seem to have a view, though: CAC bonds in the countries with heightened redenomination risk seem to be valued significantly more.

Developing Countries Issue Sovereign Debt (Lots of)

In the FT, Elaine Moore reports that “[d]eveloping economies are on course to raise a record sum on global debt markets this year, as ultra-low rates in the developed world cheapen borrowing costs for countries from Asia to South America.” By the end of the year, hard currency debt sales by countries such as Mexico, Quatar, Saudi Arabia and Argentina are expected to reach USD 125 billion.

Sovereign Debt in Bank Balance Sheets

In the FT, Martin Arnold reports about estimates by Fitch according to which

European banks would have to raise up to €170bn of extra capital or sell almost €500bn of sovereign debt if regulators push ahead with plans to break the “doom loop” tying lenders to their governments …

The European Commission and the European Central Bank support steps in that direction while some European governments oppose them.

Financial Transactions Tax—Stalled

In the FT, Jim Brunsden reports that the European Commission’s 2013 proposal to install a financial transactions tax has not made much progress. At least nine countries have to sign up.

The report highlights that key differences remain on how to craft exemptions from the tax, including the problem of how to shield transactions in other non-participating EU countries such as Britain. Other splits concern how to protect market-making activities by banks, and also what carveouts should apply for derivatives that are used by traders to hedge risk when they buy sovereign debt.

Argentina’s Debt Negotiations

In the FT, Chris Giles, Gillian Tett, Elaine Moore and Benedict Mander report about the negotiations between Argentina and the country’s creditors that are about to start, now that the new government has taken office.

Argentina’s finance minister has announced that the country intends to honor the face value of outstanding debt but wishes to negotiate interest payments.

As a sign of support from the international community, Jack Lew, Treasury secretary, announced that the US had ended its formal opposition to the World Bank and other multilateral development banks’ lending to Argentina.

Observers expect that the IMF will soon be involved to provide technical assistance.

In an FT blog, Charles Blitzer argues that successful negotiations should start with a non-disclosure agreement. He links to the Institute of International Finance‘s Principles for Stable Capital Flows and Fair Debt Restructuring.

Quasi-Sovereign Debt

In the FT, Elaine Moore and Jonathan Wheatley report about the increasing importance of sovereign-backed corporate and other debt in emerging markets.

New figures from JPMorgan and Bond Radar show that issuance of quasi-sovereign bonds outpaced that of sovereign bonds in emerging markets last year, raising the stock of such debt from $710bn in 2014 to a record $839bn by the end of 2015. By comparison, the stock of all external emerging market sovereign debt stood at $750bn at the end of last year, according to JPMorgan.

Quasi-sovereign borrowers include firms that are owned in large parts or controlled by the state, as well as local governments. Although the liabilities of these borrowers may be explicitly or implicitly guaranteed by the state, the official public debt statistics typically do not account for them.


Sovereign Debt Seniority

In a Vox column, Matthias Schlegl, Christoph Trebesch, and Mark Wright document an implicit seniority structure of external sovereign debt: IMF > Multinational > Bonds > Bilateral > Banks > Trade Credit (see the figure).

They argue that Greece’s recent default on the IMF constitutes an outlier.

… Greece in 2015 is clearly an outlier case, having defaulted on the most senior creditor (the IMF), while continuing to service historically more junior creditors. The evidence also suggests that the Eurozone rescue loans, which are essentially bilateral (government-to-government) credit, are likely to be a junior creditor class going forward. The evidence also rationalises why Greece may have an interest in exchanging the debt it owes to the IMF and the ECB into loans to the European Stability Mechanism, which is likely to be junior debt in the future, as discussed in the run-up to the July Eurozone summits. Policymakers should be aware of the associated changes in seniority and repayment incentives.

trebesch fig3 7 aug

Why Do Sovereigns Repay External Debt?

In a Vox blog post (that complements another post on Greece), Jeremy Bulow and Ken Rogoff review the academic discussion on a long-standing question—why sovereigns repay their external debt.

Bulow and Rogoff distinguish between

[t]he ‘reputation approach’ pioneered by Eaton and Gersovitz (1981) which builds on Hellwig (1977);
and the ‘direct punishments’ bargaining-theoretic approach of Bulow and Rogoff (1988b, 1989a) which in turn builds on Cohen and Sachs (1986).

They argue that the latter approach—attributing enforceable rights in foreign country courts to creditors—better explains observed outcomes.

[The] direct punishment/bargaining approach lends itself very naturally to incorporating moral hazard; …
reputation models suggest [counter factually] that the governing law of the debt is irrelevant;
[i]n standard reputation for repayment models, write-downs are decided unilaterally—creditors’ particular concerns do not really matter; …
[t]he interests and welfare of unrelated third parties does not matter in standard reputation models; …
[r]eputational debtors borrow in bad times and re-pay in good times, for purposes of income smoothing; [d]efaults, if they are to take place, occur in good times … In reality, many countries borrow as much as they can whenever they can. … Debt crises occur when countries do badly and creditors decide they want to reduce their loan exposure. To some extent, this issue can be addressed by assuming that income shocks are permanent and not transitory, but it remains difficult to rationalise country borrowing only on the threat of lost consumption smoothing. …
[c]reditor identity doesn’t matter; …
[u]nder … general assumptions, the existence of [the option to put savings abroad] leads to the unravelling of any purely reputational equilibrium.

Bulow and Rogoff add that

[a]nother important issue … is that in practice, sovereign debt renegotiations focus very much on the flow of repayments, and much less on how the stock of debt evolves. This is precisely because all sides realise that any future promises can be renegotiated.

Quantitative Easing by the ECB

The ECB announced the long-awaited expansion of asset purchases. The press release lists these main points:

  • ECB expands purchases to include bonds issued by euro area central governments, agencies and European institutions
  • Combined monthly asset purchases to amount to €60 billion
  • Purchases intended to be carried out until at least September 2016
  • Programme designed to fulfil price stability mandate

Less expected is the arrangement for the sharing of “hypothetical losses”. The ECB will directly be exposed to only 20% of the risk of the additional asset purchases.

Another ECB website provides an overview over the ECB’s open market operations.

Pari Passu and Collective Action Clauses: The New World

An IMF staff report published in September and entitled “Strengthening the Contractual Framework to Address Collective Action Problems in Sovereign Debt Restructuring” discusses recent legal developments of relevance for sovereign debt markets and implications for the sovereign debt restructuring process.

The New York court decisions (NML Capital, Ltd v. Republic of Argentina) have rendered a holdout strategy more likely to succeed. This tends to exacerbate collective action problems and raises the risk of more protracted debt restructuring processes. Market participants, including the International Capital Markets Association (ICMA) are discussing contractual clarifications and modifications in response to this challenge. The IMF observes these discussions and supports the preliminary results.

The New York court decisions established a broader interpretation of the standard pari passu clause in sovereign debt contracts. Specifically, they extended the standard notion of “protection of a creditor from legal subordination of its claims in favor of another creditor” to the broader notion that a sovereign must pay creditors on a pro rata basis. The court decisions prohibited Argentina from making payments to holders of restructured bonds unless it paid holdout creditors on a pro rata basis, and it prevented banks from making payments on Argentina’s behalf. In this context, the decisions also interpret the U.S. Foreign Sovereign Immunities Act. The scope of the rulings is not clear, not least because the decisions also refer to Argentina’s “course of conduct.” If interpreted broadly, the court decisions change the legal framework and are likely to complicate the restructuring of New York law-governed debt contracts (while probably not affecting London law-governed contracts).

Box 1 of the report discusses in detail the history of the Argentine litigation in the U.S. The report also contains an annex on the history of pari passu clauses in New York law-governed sovereign debt contracts.

Sovereign issuers have already reacted to the court decisions, by modifying the pari passu clauses in debt contracts. Also, ICMA has proposed a new standard pari passu clause, emphasising equal ranking as opposed to pro rata payments.

Collective action clauses enable a qualified majority of bondholders (e.g., 75%) of a specific bond issuance to bind the minority to the terms of a restructuring agreement. If collective action clauses operate on a series-by-series basis rather than on the total stock of debt then a blocking minority can more easily be formed and a strategy of holding out is more likely to succeed, in particular in light of the recent New York court decisions. The possibility to aggregate claims across bond series for voting purposes works in the opposite direction. Some countries have included aggregation clauses in the debt contracts, and the ESM treaty requires standardised aggregation clauses (“Euro CACs”) in Euro area government bonds as well. These clauses feature a “two limb” voting structure, requiring a majority of bondholders in each series and across all series but a lower quorum (e.g., 66%). Currently, “single limb” procedures are being discussed. These would solely require a majority across all series. To prevent abuse, such single limb procedures would have to be accompanied by safeguards that ensure inter-creditor equity, in particular a restriction to offer all affected bondholders the same (menu of) instruments. (Offering the same (menu of) instruments would generally imply that some creditors suffer larger restructuring losses than others, depending on the type of instruments they held initially. But already today, this is common and generally accepted.)

Box 2 of the report discusses the history of collective action clauses. Box 3 of the report discusses disenfranchisement provisions. Their purpose is to limit the risk of a sovereign manipulating voting processes by influencing votes of entities under its control.

Sovereign Debt Issuance under Domestic Law

In the second of his Munich Lectures in Economics, Kenneth Rogoff discussed financial crises. (His first and third lecture covered a proposal to phase out cash, see my post.)

In addition to reviewing his work with Carmen Reinhart, Rogoff returned to an earlier proposal (Bulow and Rogoff 1990, Journal of Economic Perspectives) according to which sovereigns should issue public debt under domestic law. This would avoid complications in an eventual debt restructuring process but would also make it harder to issue debt in the first place.

Rogoff reminded his audience that not only developing countries are and were subject to IMF programs; advanced economies were too (e.g., the UK in the 1950s).

“Debt Maturity Without Commitment,” JME, 2014

Journal of Monetary Economics 68(S), December 2014. PDF.

How does sovereign risk shape the maturity structure of public debt? We consider a government that balances benefits of default, due to tax savings, and costs, due to output losses. Debt issuance affects subsequent default and rollover decisions and thus, current debt prices. This induces welfare costs beyond the consumption smoothing benefits from the marginal unit of debt. The equilibrium maturity structure minimises these welfare costs. It is interior with positive gross positions and shortens during times of crisis and low output, consistent with empirical evidence.

International Bankruptcy Law back on the Agenda?

Elaine Moore writes in the FT that the idea of an international bankruptcy law gains traction, after IMF suggestions in the early 2000s for a “Sovereign Debt Restructuring Mechanism” failed to receive sufficient support. In September, 124 UN General Assembly members supported the proposal to develop a legal framework for restructuring sovereign debt.

This proposal complements the agreement among market participants, international organisations and regulators (against the backdrop of Argentina’s latest default) to change bond clauses with the intention to facilitate renegotiation through binding majority decisions.

Sovereign Debt Composition in Advanced Economies

S. M. Ali Abbas, Laura Blattner, Mark De Broeck, Asmaa El-Ganainy and Malin Hu report in Vox about their debt structure database spanning the period 1900–2011 and covering Australia, Austria, Belgium, Canada, France, Germany, Ireland, Italy, the Netherlands, Spain, Sweden, the UK, and the US. Data is disaggregated along the following dimensions: Currency; maturity (of local currency debt); marketability; holders (non-residents, national central bank, domestic commercial banks, rest).

Their main findings are:

  • Advanced economies’ debt typically was denominated in local currency, with the exception of post-WWI France and Italy.
  • Governments issued longer-dated paper in good times.
  • The share of central government debt that was issued in the form of marketable securities declined until after World War II and increased again in the 1970s, to around 80% today.
  • National central banks and domestic commercial banks held about 30% of the debt until 1970. Afterwards, non-resident participation in sovereign debt markets soared.
  • Large increases in debt often were accompanied by a rise of short-term, foreign currency-denominated debt held by the banking system, with the exception of the 1980s and 1990s where more long-term local-currency marketable debt was issued.
  • Evidence for financial repression in combination with inflation after World War II.

They suggest that countries mainly followed two strategies to reduce debt quotas. One, based on fiscal consolidation and moderate inflation, going hand in hand with long maturities. The other, based on high inflation and reliance on debt holdings by captive domestic investors, going hand in hand with shorter maturities.

Link to the data.

Banking Union

Willem Buiter, Ebrahim Rahbari and Antonio Montilla provide a detailed assessment of the need for a Euro Area banking union and the progress towards it, in a Citi Research document. Some excerpts from the abstract:

… a single supervisor, a common resolution mechanism, including a joint recapitalisation back-up, and an effective lender of last resort – is necessary for the euro area (EA) to survive.

The CA’s [comprehensive assessment’s] conclusion will likely boost EA financial conditions in coming months. Even so, we believe the CA should have been more stringent, current backstops are still inadequate, and the CA will not eliminate divergences in financial conditions …

Remaining elements of narrow banking union are also very important — These are the Single Supervisory Mechanism (SSM) and the Single Resolution Mechanism (SRM). In particular, the SRM and its bail-in provisions should materially reduce the likelihood that an EA sovereign will be dragged into insolvency through tax-payer-funded bank bailouts. Reduced moral hazard will also likely lower the likelihood and severity of future banking crises. And the combination of CA, SSM and SRM are also likely to mean that the ECB will be an effective lender of last resort for EA banks (and sovereigns).

… one key fragility remains: excessive two-way links between national sovereigns and banks. Risk-weighting of sovereign debt and concentration limits on sovereign debt holdings by banks are necessary to break these links.

A single deposit guarantee scheme is not necessary for monetary union and requires a deeper fiscal union than the minimal common backstops required to make monetary union work. It is therefore unlikely in the foreseeable future, in our view. An EA sovereign debt restructuring mechanism (SDRM) may be necessary to handle legacy sovereign debt restructurings and possible future sovereign insolvencies, but beyond the limited mutualised fiscal backstops necessary for banking union and the SDRM, deeper fiscal union is neither necessary for EA survival nor likely, for political reasons, in the foreseeable future.

The report contains many interesting figures, for example on the exposure of banks to their domestic governments; the correlation between sovereign and bank CDS spreads; lending standards; the share of bank loans in banks’ balance sheets etc.

Foreign-Law Sovereign Debt

The Economist discusses the consequences of issuing sovereign debt under foreign law. Investors in Greek government debt did better if they owned paper governed by English law—these series escaped the retroactive addition of collective-action clauses that Greece added to its domestic law bonds in 2012 before renegotiating with its creditors. For Argentine debt the situation may be reversed, due to the New York court decision in the case of Elliott Management against the Republic of Argentina. Indeed, default risk might be lower for Argentine domestic-law bonds.

Haircut Estimates for 180 Sovereign Debt Restructurings

Juan Cruces and Christoph Trebesch have posted data on 180 sovereign debt restructurings between 1978 and 2010. They analyse this data in a paper forthcoming in the American Economic Journal: Macroeconomics.

Cruces and Trebesch consider distressed restructurings of medium and long-term public or publicly guaranteed debt with foreign private creditors (including commercial banks and bondholders but excluding official creditors organised in the Paris Club) that were ever finalized. Their preferred haircut measure varies between -10% (Brazil, February 1983) and 97% (Republic of Yemen, February 2001).