July 2020
Joseph Stiglitz and Hamid Rashid
Columbia University
THE SOVEREIGN DEBT BUBBLE: THE ROLE OF THE CREDITOR COUNTRIES
While it is generally believed excessive liquidity and leverage caused the global financial crisis, the United States and Europe responded to the crisis by injecting more liquidity into their financial systems. Under quantitative easing (QE) measures, the Federal Reserve and the ECB added more than $4.4 trillion dollars to their balance sheets during 2008-2015 to stimulate recovery.13 The QE – as designed – kept interest rates low and flooded financial markets with cheap money. Instead of boosting credit and investment in the US and Europe, the QE money chased equity and debt instruments in developing countries. The US’ holding of emerging and developing country sovereign bonds, for example, has more than doubled since the global financial crisis.14 Debt crises tend to be blamed on the debtor countries. But debt contracts are voluntary arrangements and whenever a debt goes bad, the creditor is as much to blame as the debtor (indeed, in some ways more so, as the creditors are supposed to be the experts in making risk judgements and assessing capacities to repay). Moreover, there is a natural proclivity of short termism on the part of political leaders: if they can get cheap credit, they have incentives to do so. It is the creditors who are supposed to provide a check on such behaviour. That they have often failed to do so is obvious, perhaps partly because of the long history of bailouts by governments, in the past often engineered through the IMF.
QE basically exported a debt bubble to developing countries. The bubble is popping now. If developing countries experience a prolonged recession, we may see more capital outflows – at least from those countries afflicted by Covid-19 – making their downturns even worse.
THE PRIVATE BORROWERS AND MACROECONOMIC EXTERNALITIES
We cannot prevent recurrent debt crises in developing countries and ensure debt sustainability without addressing underlying factors that affect trade and current account balances. Current account deficits imply capital inflows, and in the absence of FDI, inflows typically take the form of debt. And we cannot solve the problems of chronic current account imbalances without addressing the macroeconomic externalities of debt and how borrowing affects the real exchange rate and terms of trade of the debtor.
Even in the best of times, debt contracts are imperfect as they cannot account for all possible contingencies. Payments are fixed, in spite of the fact that the country’s ability to service that debt may be affected by a shock. The current crisis presents a systemic shock to debt sustainability, which needs a systemic response.
To achieve debt sustainability, we need to address three related issues:
a.the financial means of a government to meet its external debt obligations;
b.the availability of sufficient foreign currency to make the payment in dollars or euros, as stipulated in a debt contract;
c.macroeconomic externalities of the behaviours of private debtors and creditors.
Keynes referred to the first problem as the ‘payment’ problem, which gets worse during an economic crisis. Governments face difficult trade-offs between spending to stimulate recovery and setting aside revenue to service their debt obligations. If debt payment is prioritised, economic recovery suffers, making it even harder for the government to collect the required fiscal revenues. Raising taxes during economic downturns can only make things worse. Matters are even worse in this pandemic, both because of the severity of the economic downturn and because servicing the debt takes away money needed to curb the pandemic.
The second challenge – the ‘transfer problem’ – also worsens during a crisis. While almost all of a government’s revenues are collected in local currencies, its debt obligations are in dollars or euros, which become scarce during a global economic crisis such as the current one, where export earnings diminish quickly. The availability of foreign exchange to service sovereign debt of a government depends on factors, such as exports and remittance inflows, largely out of the control of the government.
Externalities generated by the borrowing by private sector further exacerbate the transfer problem. In 2018, 82 of the 111 countries which are the centre of discussion of this Policy Insight had private sector external debt, up from 60 in 2008. When allowed, the domestic private sector borrows from foreign creditors without any regard for externalities of their borrowing decisions – for example, that the magnitude of the depreciation of the exchange in the event of a shock will depend on how much they (and other firms) borrow. There are similar externalities associated with short-term capital inflows from abroad.
Similarly, at the first sign of a crisis, private creditors – with exposures to both sovereign and private sector borrowers in developing countries – typically rush to take out their money, without any regard for the externalities of their actions. The quick outflows of capital trigger exchange rate depreciations.
As the exchange rate depreciates, it becomes harder for the government to buy the required foreign currency and service its foreign debt obligations. An adverse dynamic can arise: a fall in the exchange rate increases both precautionary and speculative demand for foreign exchange from the private sector. The externalities of the actions of the private sector borrowers and private creditors make debt servicing more difficult and expensive for government, which must be taken into account if we are to avoid recurrent debt crisis in developing countries.
III. Calls for urgent action and standstills
Policymakers and academics around the world have joined the call for immediate debt relief for low- and middle-income countries (Berglöf et al., 2020; Okonjo-Iweala et al., 2020) as we described earlier. Much of the attention has centred around a debt standstill. There have been numerous calls to waive debt repayments this year, including $44 billion from the African countries. Others have called for a marketbased solution, activating collective action clauses (Gelpern et al., 2020) for a one- to two-year standstill on all external-debt repayments, both interest and principal. A number of G20 members have already announced debt relief for the poorest countries for a year. The IMF has cancelled six months of debt payment for 27 poorest countries, as of June 2020.
There is a strong rationale for such standstills, which apply domestically as well as internationally. While the rest of the world has been put on hold by the pandemic, financial markets have not – they continue to charge interest, and worse, those who cannot honour their obligations for serving debt may have to pay additional fees and charges. Equity demands more symmetry: as the rest of the world goes into a standstill, so should finance. One of the advantages of a debt standstill is that since the period is (assumed to be) short, the losses will be small, so the incentive to go to court rather than cooperate with a standstill will be limited.
The standstills adopted so far have, however, been of limited scope. The measures proposed by the G20 cover only 77 low- and middle -income countries, including the Least Developed Countries and countries that are eligible for the International Development Association (IDA) credits or the concessional lending window of the World Bank and the so-called ‘blend’ countries. According to OECD estimates, the debt servicing standstill proposed by the G20 would cover at most $16.5 billion that these countries would have paid to their bilateral creditors in 2020. This would be a respite, not debt relief, according to the OECD (2020). The G20 Agreement in April suggests that the standstill should be net present value (NPV) neutral, which means that the countries not paying the debt service in 2020 must repay the full amount – including accrued interests in 2020 and 2021 – over 2022-2024 (with a grace period in 2021). The accrued amount could be quite high for non-concessional loans.
The actual standstill amount could be significantly smaller, depending on the participation of the creditor and debtor countries and what debt is counted as official bilateral credit. Kenya, for example, has indicated that it would not request a debt standstill under the G20 initiative, as it would curtail the country’s access to nonconcessional loans in the international capital market. There is no call from the G20 members to restructure and reduce the debt burden of the poorest countries.
For numerous reasons – inter-creditor equity and economic efficiency – a standstill must be comprehensive, including private creditors. A standstill involving only official creditors is unlikely to work. It would not be sufficient. It would not be politically palatable. Official creditors may rightly be concerned that their standstill would do little to help the debtor country; it would simply give them the liquidity to service the debt they owe to private creditors. Furthermore, if the standstill is enforced on public debt payment only, and not on private debt payments, it will set a costly precedent. The debt of private corporations in developing countries will be deemed safer than sovereign debt, which will increase the borrowing cost of governments relative to the borrowing cost of private sector entities. It would suggest private creditors have seniority of claims over bilateral and multilateral creditors.
There is legitimate concern, however, that private creditors will not easily accept a standstill. The newly established private creditor group – the Africa Private Creditor Working Group – has already rejected the G20 call for the modest debt relief for the poorest countries in Africa.20 It is unlikely that they will cooperate on a larger-scale debt relief. The fear is that private creditors will use every means at their disposal – including legal and political coercion – to block the standstill, even if it means total disarray.
Indeed, in the absence of strong collective action clauses and trusteeship provisions (discussed below), there are incentives for some bondholders to act as holdouts in order to get a better deal for themselves. If a group of creditors voluntarily agree to a debt standstill, other creditors and bondholders may still demand debt servicing. Many sovereign bonds include cross-default clauses, which means a technical default on one bond can qualify as default on all other outstanding debt. This can trigger acceleration, with creditors demanding immediate payment of full principal amount and accrued interests. This puts enormous power in the hands of a few recalcitrant creditors.
Given that nearly two thirds of outstanding sovereign bonds are governed by New York law and given that New York courts traditionally favour creditors (especially US creditors suing foreign governments), debt standstills may end up costing developing countries millions in legal fees and pre-judgement penalties. New York state law allows a plaintiff creditor to collect a 9% pre-judgement penalty on all overdue payment of interest (Zuckerbrod, 2017).21 While the objective of pre-judgement interest is compensatory, this 9% interest rate provides an incentive for a bondholder to sue sovereign borrowers, especially when, as now, the market interest rate is significantly lower.
Similar concerns are raised about other voluntary mechanisms. A group of economists and lawyers, for instance, have proposed the establishment of a central credit facility (CCF) for each country seeking debt relief to activate a 12-month standstill on debt services to both official and private creditors (Bolton et al., 2020). The CCF would allow a country seeking temporary relief to, in effect, use deferred interest payments as emergency funds to fight the pandemic. The group proposes a multilateral institution like the World Bank to manage and monitor the appropriate use of funds that debtor governments would have otherwise used to service their external debt. Such an arrangement would be of considerable benefit, if the creditors could be persuaded to cooperate. But the evidence to date is that many may not.
A standstill, however designed and implemented, is just short-term palliative; these measures are, at most, band-aids for a much larger systemic problem. It must be viewed as a means to an end for achieving a sustainable solution to the systemic debt problem of developing countries. A standstill, even without interest accumulating, doesn’t solve the fundamental debt problem discussed earlier; and in particular it doesn’t address the fact that even debts that seemed sustainable prior to the pandemic may not be sustainable post-pandemic. The pandemic and its economic impacts will last much longer than a year. Ad-hoc and piecemeal debt relief will help delay a debt crisis but could saddle countries with even higher debt burdens. What has been done, and even what has been proposed, is insufficient.
For many developing countries, more than a standstill will be required: there is a clear need for comprehensive debt restructuring. This is true not just for the poorest countries that will be badly afflicted, but also for middle-income countries. And debt relief needs to be comprehensive, including and especially on debt these countries owe to private creditors.
It is critical, both from the perspective of the debtor countries and the international community, that there be an orderly debt restructuring. In most of the cases at issue, there will be some form of debt restructuring; the key question is: will it be orderly or disorderly, and will it sufficiently protect the interest of the debtor countries? If the countries don’t have the resources to pay, they won’t pay. It is not, as some creditors seem to think, simply a zero-sum game. The strategy of ‘too little, too late’ has proven to be negative sum, with both creditors and debtors suffering. One might conclude, therefore, that ‘rationality’ would ensure that markets, through free bargaining, would ensure an efficient outcome. But it is well known that that is not the case. We have already noted some of the externalities associated with private debt; and these are even greater in the context of debt renegotiations. Some creditors may believe that they can grab more for themselves, through the pursuit of ‘tough’ bargaining stances, even if other creditors, and the debtor, lose far more than what they can gain. In general, the outcome of unbridled bargaining is not Pareto efficient.
While a standstill doesn’t adequately address the long run debt problems, the time provided by the standstill can be used to restructure debt – significantly reducing debt burdens, while securing favourable terms, and helping to restore the balance of payment equilibrium. Ideally, such restructurings can be used to incorporate better terms and conditions for existing and new bonds – including state-contingent payment options and provisions related to trusteeship and bankruptcy procedures, such as collective action clauses. Debt restructuring typically takes an extended period of time, and it would be preferable that such restructuring be done before the country actually plunges into a crisis.
Unfortunately, as difficult as getting cooperative voluntary action in a standstill is, getting such cooperation in debt restructuring will be even more difficult, as evidenced by the ongoing and difficult negotiations that the countries currently in crisis are facing. Making matters even worse, some creditors are demanding that the newly restructured bonds contain provisions that will make future debt problems even worse (for example, weakening the collective action clauses).
IV.Achieving comprehensive debt restructurings
The previous section outlined the need for urgent and comprehensive actions that go beyond the standstills that have so far been the centre of attention, but suggested that the necessary cooperation has not been forthcoming from the private sector. Given the extraordinary challenges, developing country governments will need to deploy a combination of legal and administrative measures to activate standstills and the necessary debt restructurings. There will have to be a combination of carrots and sticks. This will require governments of debtor and creditor countries to collaborate – not just as a moral imperative, but as an economic imperative for avoiding a debt catastrophe, with its implications for peace and economic stability. It should be obvious that the developing countries and emerging markets will need to act in concert: if they negotiate bilaterally with their official and private creditors, they will likely receive insufficient relief, which will only help them postpone a debt crisis.
The creditor countries can undertake a number of actions to support standstill measures. Most of the bonds are issued in a relatively few jurisdictions (namely, under New York and English laws), so actions by these jurisdictions can make a difference. Earlier, we referred to the 9% interest on pre-judgment interest in New York that acts as a deterrent to reaching a successful debt restructuring. The Governor of New York proposed bringing the interest rate down from its current usurious level (exessive rate) to one which better reflects the intent of providing adequate compensation for deferred payments, but the provision doing so seems to have disappeared from the budget legislation as it was enacted in April 2020, presumably as a result of the influence of creditors. There needs to be concerted efforts to bring down pre-judgement penalties to minimise the holdout (agreement) problems in sovereign debt restructuring.
New York used to have a regulation (called ‘Champerty’) preventing the buying of bonds at deep discount with the intent of suing to recover full value. This is, of course, the business model of the vulture funds, which have played such a central role in preventing orderly debt restructurings. Under the influence of the vulture funds, New York State passed aprovision exempting transactions in excess of $500,000, thus effectively exempting the vulture funds.
Note: a fund which invests in companies or properties which are performing poorly and may therefore be undervalued
Usury provisions, regulating excessive interest rates, had long been part of legal frameworks around the world until such regulatory constraints came to be looked upon with disfavour in the era of neoliberalism. Domestically, some of the financial instability and predatory lending observed in recent decades has been attributed to the repeal of these usury constraints. Similarly, high interest rates have almost surely encouraged excessive lending internationally. Constraints on these interest rates would both have discouraged such lending and would have meant, if the lending occurred, the burden on the countries would have been reduced. (Research in economics in recent years has emphasised how in imperfectly competitive markets, such constraints can actually improve economic performance. There is evidence that the international credit market to developing countries and emerging markets is a far from perfect market, yielding excessive risk-adjusted returns.)
Finally, taxation may affect incentives. For instance, a surtax (additional tax) on excessive earnings (rates of return in excess of, say, 4% above the US treasury rate) and similar surcharge on excessive capital gains (say, a 7% real return per year) would discourage both highrisk lending and holdouts, though it could at the same time lead to larger decreases in bond prices in times of stress.
Perhaps the most effective way of encouraging debt restructurings in the face of a debt crisis would be to clarify the legal framework – putting into laws governing sovereign debt the kinds of understandings that are part of Chapter 9 of the US Bankruptcy Code that Congress adopted during the height of the Great Depression to protect the assets (pension funds, public assets such as hospitals, recreation centres, etc.) of local governments (municipalities) against the claims of private creditors, to ensure that public authorities could continue provide the essential public services that they provide, and to recognise the informal claimants, whose ‘claims’ may have equal or greater validity than the formal contractual claimants. Unlike the bankruptcy procedures designed for private borrowers, the Chapter 9 procedures give significant protection and power to public entities to restructure their debt in the public interest. The idea is simple: social obligations of a public entity – providing basic services to the citizens and honouring commitments to pensioners, among others – should have priority over financial claims of the private creditors.
Chapter 9 procedures guided the restructuring of $18 billion debt that the city of Detroit owed to bondholders, securing steep haircuts while protecting the city’s assets and obligations to its pensioners. There has been calls along these lines to protect debtor country’s assets from attachments by private creditors (Buchheit et al., 2013), which can significantly address the ‘holdout’ problem.27 The perspectives we have just discussed can be viewed as consistent with a number of longstanding principles reviewed in the next subsection. In presenting these, we do so not from the perspective of law, but from that of economics: they are a partial substitute for the state-contingent contracts (discussed later) that would represent a better way of risk sharing between creditors and debtors.
To be continued
To Be Continued
gandatmadi46@yahoo.com