The Atlantic Council hosted a panel on debt restructuring this week in the wake of Greece's decision to utilize its holdings of Special Drawing Rights (SDRs) at the IMF to cover its payment owing to the Fund. The panel featured professor Joe Stiglitz, Nobel laureate from Columbia University, Sean Hagan, General Counsel of the IMF, Maarten Petermann from JP Morgan, and Ambassador Cecilia Nahon, Argentina's representative in Washington. The subject of sovereign debt restructuring is once again on the international policy agenda, the moderator argued, because of the high-profile cases of Argentina, Greece and, more recently, Ukraine. This really is, it seems, the year of restructuring.
As Maarten Petermann argued, debt restructurings are necessarily messy (otherwise there may be too many in equilibrium, resulting in a contraction in the market for sovereign lending). The question, I’d argue, is whether there are deadweight losses to the sovereign borrower, creditors and the international community more generally that could be reduced through governance reforms that promote timely, orderly restructurings. A decade ago, the IMF under then Deputy Managing Director Anne Krueger proposed a formal statutory approach — the Sovereign Debt Restructuring Mechanism (SDRM) — to reducing these deadweight losses. Sean Hagan pointed out that the necessary political support for the SDRM did not exist then and does not exist today. The SDRM was put on the proverbial “back burner” and efforts were made to improve the debt restructuring process through so-called voluntary approaches.
Professor Stiglitz explained why reliance on voluntary approaches is likely to be insufficient. There have been three big changes in the sovereign bond market over the past two decades, he argued, that have changed the rules of the game or rules of engagement of debt restructuring. The first change is the increase in heterogeneity and number of creditors in contrast to the 1980s. Thirty years ago, lending to sovereign borrowers was dominated by commercial banks; the switch to bonded debt has resulted in a far larger universe of investors. The second change in the sovereign debt market is the development of Credit Default Swaps (CDS), which allow investors to buy insurance against default. The third factor changing the rules of the game is a New York judicial ruling in support of certain holdout investors.
Each of these changes has increased the difficulties of a successful restructuring. The increased number and greater heterogeneity of investors, for example, has increased the coordination challenges of getting an agreement. Consider a hypothetical debt restructuring in which there is a stock of debt with net present value, D, the sovereign has a maximum (net-present value) debt-servicing capacity of S, and the number of creditors with identically sized claims is n. In a frictionless world, in which there are no negotiating costs to getting an agreement and the sovereign can credibly commit to S, the haircut borne by each creditor is (D – S)/n.
With perfect information on both sides of the negotiation, this restructuring would emerge more or less automatically.
But what happens if there are cost to negotiation that vary with the number of investors because, say, of the difficulty of getting an agreement? The haircut would then be determined in part by the cost function, which might look something like C = c∙n. The size of the haircut would then be (D – S – c∙n)/n; the greater the number of investors, the larger the haircut. It is also possible that debt-servicing capacity is negatively affected by the number of creditors. This effect might arise if economic dislocation increases the longer a restructuring is delayed. In this case, the size of the haircut is: (D –S∙n – c∙n)/n. Of course, the cost function could be an increasing function of the number of investors, so that C = cn.
All of this changes the size of the haircut required. It does not necessarily imply that restructurings will be more difficult to achieve, particularly if all investors have the same information. This is where heterogeneity and imperfect information come into play. If different investors are of different size and have different objectives, the incentive to come to an agreement may differ. Consider three investor types. First, large investors with a larger fraction of wealth exposed to the potential default of the sovereign, who may want to agree more quickly. Second, investors holding a small share of the outstanding debt stock and with a relatively small exposure who may be prepared to delay, hoping that other investors will settle early, increasing their expected return. And, third, investors prepared to buy up the claims of other investors, providing liquidity to distressed assets, to litigate in order to "hold out" for a smaller (or no) haircut.
The development of CDS instruments has, arguably, increased the relative influence of the second investor class — with insurance against default, these investors have less of an incentive to agree to a timely restructuring. Indeed, incentives may be sufficiently distorted that they are made better off in the event of default, at which point their CDS contracts are exercised. This possibility presents a clear moral hazard problem, in that such investors can increase the probability of default by demonstrated intransigence in the negotiations by demanding debt-service payments greater than S. The threat of moral hazard explains the use of co-insurance in insurance contracts to align incentives and prevent strategic behavior on the part of the insured.
The returns to the third investor class, meanwhile, have likely increased as a result of changes to the doctrine of Champerty. Formerly, this legal doctrine denied legal protection to investors who bought debt solely for the purpose of litigating for payment. At the same time, a judicial interpretation of pari passu in an Argentine bond issue that was restructured in the wake of default a decade ago. What the ruling does is marries a judgement with a means of enforcement by threatening to impeding payments on previously restructured bonds. With the Champerty defense weakened and the potential returns to holding out increased, the prospect of a return to the lengthy sovereign debt restructurings that were typical a decade and more ago looms large.
In this respect, the voluntary restructurings of the past decade are regarded by many as relatively timely and efficient. And, as Maarten Petermann noted, the speed of the process is determined by the exchange offer: if a distressed sovereign seeks a timely restructuring and early return to the capital market, it can offer a generous debt exchange that secures a high participation. But what is fair to creditors may be prejudicial to others. In the words of Stiglitz, we can't just say the "process works" merely because there are restructurings. The outcomes of these restructurings have to be evaluated on the basis of equity as well as efficiency (as defined, say, by the timely return to capital markets). If the claims of other creditors — such as domestic pensioners — are violated, the outcome is not optimal.
In fact, in the words of the IMF, the problem of sovereign debt is that restructurings are frequently “too little too late.” The increased leverage wielded by holdout investors resulting from the New York court’s interpretation of pari passu may exacerbate this problem. Sean Hagan noted recent innovations in collective action clauses (CACs) are designed to address the potential threat. He observed that revised CACs are not a panacea, however, as they do not address the roughly $900 billion in outstanding bonds. Moreover, as I have suggested, here, developments over the past several decades illustrate, bonding "technology" evolves; imaginative legal minds will likely find means to circumvent even the most clearly constructed CACs.
Hagan also made a crucial observation. A key issue in sovereign debt restructuring, he said, is discriminating between insolvency, in which a reduction in the net present value of the debt is required, versus illiquidity. Doing so is straightforward in the case of corporate bankruptcy; not so in the case of sovereign insolvency. As a result, the IMF is often on the horns of a difficult dilemma when assisting a sovereign in financial distress. Lending into insolvency hurts the country and hurts private creditors through subordination. Debt is debt, regardless of whether it is owed to a private creditor or the IMF. (As Doug Purvis taught me, lo these many years ago, foreign debt is a claim on a share of country output). Private creditors are affected by the subordination of their claims resulting from the IMF’s preferred creditor status — or, the convention that the IMF gets repaid in whole before private creditors get anything.
As Hagan noted, "the Fund's lending policies are at the core of the timing issue." By lending into financial crises, the IMF could reduce the incentive for private creditors to engage in a timely restructuring. Why would they, if it is likely that the IMF will provide the exit ramp they need to get out whole? There is a risk, however, that by not lending, the crisis in one country spills over to neighbouring countries, with adverse effects for the global economy writ large. That outcome is inconsistent with the IMF’s “prime directive” of promoting international financial (monetary) stability so that its members eschew “policies destructive of national and international prosperity.” Fund lending decisions are complicated by the fear of contagion. In short, progress on sovereign debt restructuring may therefore hinge on the IMF stepping away from the "systemic exemption" under which it can lend even where there are serious solvency concerns. Such an initiative, Hagan suggested, would be "a limited change, but a substantial change."
That strikes me as exactly right. But there is a problem of dynamic inconsistency: absent a framework for the timely, orderly restructuring of private claims, the IMF and its shareholders may be reluctant to withhold financing, fearing the consequences of possible contagion, even if financing harms the borrower and the private creditors who remain. In this respect, the absence of clear rules of the game creates uncertainty that magnifies the costs of restructuring. Getting the clarity and consistency that is required — whether through “soft law” in the practices of the IFIs or in more formal statutory approach — should be a priority for the international community. Doing so would help hasten the end of the New Age of Uncertainty.