Sovereign Analysis /
Sri Lanka

Sri Lanka's debt restructuring explained

  • We recently took a deep dive into Sri Lanka's debt restructuring on NDB Securities’ Rational Investor Podcast

  • Topics include lessons from past restructurings, the debate on including domestic debt, and other key challenges

  • We provide the full transcript below, alongside a link to recording

Sri Lanka's debt restructuring explained
Tellimer Research
23 November 2022
Published byTellimer Research

This is the transcript of a recent appearance on Colombo-based broker NDB Securities’ Rational Investor Podcast entitled “Sovereign Debt Restructuring 101, and the Road Ahead for Sri Lanka”, moderated by NDBS’s Head of Research Vajirapanie Bandaranayake. It has been edited for clarity with relevant backlinks added.

The full podcast can be viewed/listened to here or on your favourite streaming platform.

Moderator: There has been a wave of sovereign debt defaults across the globe in recent times. What is the reason behind this phenomenon?

Patrick: Many emerging and frontier market economies were facing rising debt vulnerabilities in the run-up to the pandemic. The average debt burden for emerging markets rose from 38% of GDP in 2013 at the time of the 'Taper Tantrum' to 55% of GDP in 2019. Then, due to the need for post-Covid fiscal stimulus, fiscal deficits widened by a cumulative 6.7% of GDP among emerging markets on average from 2020-22. This pushed up debt even further to 66% of GDP by 2022, nearly double what it was a decade ago. And then, of course, we entered this year with higher debt burdens and worsening fiscal dynamics, and economies were hit by another series of shocks caused by Russia's invasion of Ukraine, high global inflation, rising commodity prices, and tightening global financial conditions. At the same time, rising yields have locked many emerging economies out of the market. So, those with large external financing needs – primarily commodity importers – are having trouble financing their balance of payments deficits. As such, they are looking at alternative financing sources like the IMF and ad hoc support from bilateral partners to plug those gaps. So, it has really just been a unique moment where there has been a series of crises all converging at a time when debt vulnerabilities were already on the rise for the past decade or so.

Moderator: Now I'm going to narrow this discussion down to Sri Lanka. What is the usual process for debt restructuring; for both external and domestic debt?

Patrick: Sri Lanka is already a couple of steps into the process. Initially, it starts with the communication of the intention to restructure, which happened back in April 2022. The next step is generally to reach out to the IMF to negotiate a recovery programme and define the restructuring envelope and parameters. Once there is some clarity on the reform programme, that materialises into a staff-level agreement, which Sri Lanka received in September 2022. So, at this point, the next step is to begin implementing ‘prior actions’, which are basically policy prerequisites that allow the IMF to approve the programme at a Board level and to disburse financing. The IMF will also require what they call ‘financing assurances’ from Sri Lanka's bilateral creditors, which include China, India, and Japan, among others. Only once those financing assurances are received, the board will be able to approve the IMF programme and begin disbursing funds.

At that point, discussions with private creditors can really start to take off in earnest. They will be expected to provide debt relief on comparable terms to what's been hashed out by official creditors earlier in the process. Once there is an agreement across all of Sri Lanka's bilateral and commercial creditors on the parameters and terms of the relief, the restructuring can be finalised and integrated into the broader reform programme. So, Sri Lanka is already a couple of steps into the process, but still has quite a long way to go.

Moderator: So, if I have understood it right, the IMF programme is contingent upon receiving financing assurances from the bilateral creditors and also showing ‘good faith’ to private creditors too. Is that right?

Patrick: That's right, and the process has taken quite some time in other countries. We have ongoing restructurings in Zambia, Ethiopia, and Chad under the Common Framework, among others. In Zambia's case, for example, it took c6.5 months after the staff-level agreement for the official creditor committee to finally meet for the first time, and then another 2.5 months for IMF Board approval. So, from that perspective, Sri Lanka is still quite early in the process.

Moderator: Sri Lanka has already begun discussions with the official creditors, and a few weeks ago, the Central Bank of Sri Lanka, together with the Finance Ministry, had a presentation to the creditors. The government expects to finalise the IMF programme by end of the year. Is this possible?

Patrick: I think the timeline that the government laid out in their recent investor presentation is an ambitious one. They want to finalise the IMF programme by end of the year and the restructuring itself by mid-2023, at the latest. As I said, taking Zambia as our precedent, the process took quite a bit longer than that. So, if it takes nine months to get Board approval and get financing assurances, that already puts Sri Lanka into mid-2023. And it's worth noting that the restructuring is likely to be quite a bit more complicated in Sri Lanka. There is a much more diverse mix of creditors and types of instruments that need to be dealt with. So, from that perspective, it could even take longer. That said, if the recent financing assurances that were provided in Zambia are symbolic of greater urgency from the official sector to move the process forward, then my pessimism might be misplaced. It comes down to how quickly Sri Lanka's official creditors are willing to sit down at the table, form a committee, and provide those financing assurances. At this stage, they are really in the driver's seat and the process will go as quickly or as slowly as they are willing.

Moderator: In terms of forms of relief, what are the typical menu options for debt restructuring; for both local and external debt? And maybe this is a little premature question to ask, but what do you think the most likely scenarios for Sri Lanka would be?

Patrick: I think there are two key categories when you are looking at menu options for Sri Lanka. The first pertains to the restructuring envelope or parameter, which basically means the debt that will be included or excluded from the restructuring process.

At a very basic level, this comes down to whether external or domestic debt be included (or both). When you are looking at an external restructuring, bilateral and commercial external debt will be included, but most multilateral debt, which is generally viewed as super-senior, will not be. Other options include whether or not state-owned enterprise (SOE) debt will be included, whether guaranteed versus non-guaranteed debt will be included, and, if the domestic debt is included in the process, whether it will be T-bills or just bonds, and of course in Sri Lanka’s case how Sri Lanka development bonds (SLDB) will be treated.

After the envelope is determined, there are a number of menu options for how to restructure the debt. The first is nominal haircuts, which are reductions to the face value of the principal. On the other end of the spectrum, there is what we call cash flow relief, which is either maturity extensions, interest grace periods, coupon cuts, or other esoteric coupon structures like step-ups. On top of this, the government may have the option to throw in sweeteners like GDP warrants, for example, which is what we saw in Ukraine's recent restructuring to help induce creditors to take larger upfront haircuts or cash flow relief.

Different creditors will have different preferences. Some, like China, for example, prefer not to take any upfront nominal haircuts and instead offer cash flow relief. Other creditors may be more willing to provide nominal haircuts in exchange for a more 'normal' coupon structure. But ultimately, the key point is that relief needs to be equal in present value terms across all available options. There may very well be a menu of different options that creditors can choose from as long as it satisfies that equal treatment requirement.

Moderator: In your opinion, what precedent has Zambia set for countries like Sri Lanka? The IMF granted the board-level approval for Zambia for a US$1.3bn extended credit facility in early September 2022, together with US$8bn cash flow relief. They published a debt sustainability analysis (DSA) that serves as the basis for restructuring. As per the DSA, a 35% to 45% of net present value cut is being looked at [Note: This has since been clarified to be 49%].

Patrick: I think Zambia is setting all sorts of interesting precedents for how debt restructuring is conducted post-Covid, especially for countries like Sri Lanka that have a large amount of debt to China. I think it's worth noting that Zambia's restructuring is taking place under the Common Framework, but as a middle-income country, Sri Lanka is not currently a part of that process. That said, the principles that apply to the restructuring are still likely to be the same. So, Zambia is an interesting comparator for Sri Lanka in the sense that it has a similar debt level, similar debt service burden, and similar composition between domestic and external debt. One interesting part of Zambia's restructuring is that the IMF said in their latest report that domestic debt will not be included in the restructuring, because it's around one-third of banking sector assets, and there will be financial stability concerns if it is included. In Sri Lanka, domestic debt is almost identical as a percentage of banking sector assets and almost identical relative to the total debt stock and debt service burden.

However, I think excluding domestic debt in Sri Lanka is likely to be more controversial. One reason for this is that in Zambia external creditors hold cUS$3.25bn – or over one-quarter – of the domestic debt stock. So, they are unlikely to push back against the government and IMF’s plans to exclude it from the restructuring. Whereas in Sri Lanka, the amount of non-resident holdings of domestic debt is negligible, and excluding domestic debt from the process means that heavier relief will be required from Sri Lanka's external creditors to meet those same debt sustainability thresholds. So, this is just one of a few aspects that are likely to be controversial in Sri Lanka and potentially differ from the restructuring in Zambia.

Moderator: Right. Domestic debt restructuring; we'll come to that a little bit later because I have a whole lot of questions on that. Now, let’s discuss ‘Equal Treatment’ in debt restructuring. This is a topic that has been widely discussed among policy circles here lately. I am trying to understand what is equal treatment and, practically, are all debts treated equally?

Patrick: Equal treatment is really the cornerstone principle of any external debt restructuring. At a basic level, all bilateral and commercial creditors involved in the restructuring must provide equal treatment in present value terms. Now, in practice, this is really difficult to calculate because the present value of relief depends on the discount rate that you choose, which is a topic we can explore later if you would like. But the general principle is that relief from one set of creditors should not be used to pay for less generous relief provided by another set of creditors, thus creating a free rider problem.

Moderator: Now let’s talk about Sri Lanka’s International Sovereign Bonds (ISBs) in light of the Equal Treatment principle. The total ISB stock is cUS$12bn, of which nearly 30% is owned by the local banking sector. Given the banking sector’s sizeable exposure to the ISBs, and the likely challenges to the financial system’s stability from the restructuring of ISBs, what would be the ideal scenario; discrimination among identical instruments, or equal treatment among identical instruments, followed by additional compensation such as banking sector recapitalisation?

Patrick: In my view, there is absolutely no space for differing treatment within instruments, especially as it pertains to Sri Lanka's eurobonds. Of course, there is a risk that restructuring ISBs causes some financial stability problems for the banking sector, but ultimately the spillover effects will need to be analysed in advance via stress tests on domestic banks, and then managed via regulatory forbearance and if needed, recapitalisation. But there is absolutely no room within the context of the current restructuring framework for differing treatment across holders of the same debt instruments.

Moderator: Okay, now let’s talk about Sri Lanka Development Bonds in light of the Equal Treatment principle. Sri Lankan authorities continue to rule out the possibility of restructuring SLDBs. These are instruments that are denominated in foreign currency, issued under domestic law, and held by residents. What we hear is that for some of the maturing SLDBs, the government has already settled the creditors in local currency and regulatory forbearance has been given in terms of managing the net open position.

Patrick: I think the exclusion of SLDBs is also going to be a non-starter for private bondholders. The reason is that for every instrument or class of creditor that is excluded, there is more relief required from the creditors that do participate. So, after excluding super-senior multilateral debt from the restructuring, only c75% of Sri Lanka's external debt stock is ‘restructurable’, so to speak. If SLDBs are excluded – which are c3.5% of GDP, and held almost entirely by residents – then the ‘restructurable’ external debt stock falls even further to only around two-thirds of the total, which will require an even heavier lift from bilateral and private external creditors. So, given that SLDBs are a foreign currency-denominated instrument, regardless of the law under which they were issued, I think private bondholders will push hard for their inclusion in the restructuring. I appreciate the point that there is already some relief being offered by settling obligations in LKR or by extending maturities, but ultimately in present value terms it is going to be quite a bit smaller than the relief that's being offered by Sri Lanka’s other external creditors. I think they are going to push hard for the principle of equal treatment to apply to the SLDBs as well.

Moderator: Now let's talk about the possibility of a local currency debt restructuring happening here in Sri Lanka. Currently, many economists have put forward the assumption that only the central bank’s balance sheet will get restructured. What are your views on this, and restructuring only the central bank’s balance sheet; is it enough to move the needle?

Patrick: I think this is going to be a central question in Sri Lanka's debt restructuring. Zooming out a bit and taking a historical perspective, since 1980 domestic debt has been directly restructured in only c30% of all sovereign debt restructurings. But there have been about another third of restructurings where it has taken place by the back door, so to speak, via financial repression and inflation. And that's what we're seeing in Sri Lanka right now. With the country’s inflation running at 70% and a 45% currency devaluation year-to-date, domestic bondholders will argue that they have already experienced a pretty large upfront haircut. Just looking at the impact of the devaluation and inflation, local currency debt has fallen from c61% of GDP at the end of last year to 52% in mid-2022 and from c53% to 43% of the total, which means that there is less of a strong argument than there was last year for domestic debt being included in the process.

That said, Sri Lanka's debt burden is incredibly high, and excluding domestic debt from the process is going to make it difficult to restore debt sustainability. As I mentioned earlier, Zambia's restructuring sets an interesting precedent, and the IMF determined that the cost of including domestic debt in the restructuring process is likely to outweigh the benefits due to financial stability risks. In Sri Lanka, which has a similar portion of banking sector assets in government debt and has a similar debt stock and composition as Zambia, presumably that argument would hold as well.

We have calculated that due to limited excess capital among Sri Lanka's five largest banks, recapitalisation would likely be needed once domestic debt haircuts exceed c5% without any regulatory forbearance and c14% with regulatory forbearance. If the government has to step in to recapitalise banks, then the fiscal cost of doing so could offset the savings from the debt restructuring.

That said, domestic debt absorbs well over three-quarters of Sri Lanka's interest burden, which is unsustainably high. As I said, excluding it from the restructuring is going to make it even more difficult to hit those debt sustainability thresholds. So, I think Sri Lanka's bondholders will argue forcefully for its inclusion in the restructuring.

Moderator: So what you are saying is restructuring only the central bank’s balance sheet is not enough in terms of achieving the debt sustainability targets, is it?

Patrick: It would certainly help on the margins but, at the end of the day, the Central Bank of Sri Lanka holds only c20% of domestic government securities. So, restructuring just the central bank’s debt would have a noticeable, but still quite a small impact on the overall debt stock. Including the central bank’s securities could still negatively affect its income and capital ratios and its pool of marketable instruments to conduct open market operations. In an extreme situation, it could even require recapitalisation. So, just looking at the central bank’s debt does not necessarily forego the problems of including domestic debt in a more holistic way, and only limits the pool of ‘restructurable’ securities.

Moderator: If there is a local currency sovereign debt restructuring, will the debtholders be treated equally, or will the incidence of a domestic debt restructuring be disproportionately shared among the creditors? I am asking this question because most of the local sovereign debt is held by the local banking sector and the Employees’ Provident Fund. Also, what inferences can we draw from the countries that have gone through a domestic debt restructuring, such as Barbados and Suriname?

Patrick: Yes, there is much more leeway on the principle of equal treatment when it comes to domestic debt restructuring, which is ultimately determined under domestic law. Treasury bills are often excluded from the process. That said, they are not always excluded. If the financial stability considerations warrant a more lenient treatment for certain types of domestic creditors, and/or if the cost of fiscal support or recapitalisation outweighs the benefits of including certain creditors, then treatment could differ across different types of domestic creditors.

Looking at some historical examples, Jamaica did a domestic debt restructuring in 2010 and 2013, which involved coupon cuts and maturity extensions that led to NPV relief of c10% to 20%, but T bills were excluded from the restructuring. In 2018, Barbados restructured both domestic and external debt, and included T bills in the restructuring because they were a very large portion of the domestic debt stock. Similar to Jamaica, Suriname excluded treasury bills. So, there is a bit more leeway on the equal treatment principle when it comes to domestic debt restructuring, and it really comes down to balancing that financial stability risks and recapitalisation costs with the debt reduction benefits across different instruments and classes of creditors.

Moderator: Let us now dig a bit deeper into the restructuring of treasury bills. Sri Lanka's total domestic government securities (both bills and bonds) are cLKR12tn, of which treasury bills account for cLKR3.8tn (c30% of the entire debt stock). Do you think treasury bills will be included in the restructuring process? The reason I am asking this is there is a notion among local investors that T bills will not get restructured. As a result, we have seen that in the recent treasury bill auctions, it was the shorter end of the yield curve that got oversubscribed repeatedly.

Patrick: Yes, that is a good question. I think it is a bit too early to speculate at this point. But, just by way of example, in Barbados T bills were included in the restructuring, and T bills were c40% of the domestic debt stock. So, a bit higher than in Sri Lanka, but not completely out of the ballpark. While T bills are less than one-third of the domestic debt stock, the interest rates on those instruments are quite high, c30%, so they constitute a relatively high portion of the debt service bill. If Sri Lanka is able to roll those over at lower rates, then maybe there is no need to include them in the restructuring. I think the government and the IMF will strive to exclude them as much as possible. But it really comes down to the debt sustainability analysis, and if that says sustainability cannot be restored if T bills are not included in the restructuring, then they may have to be looped into the process as well.

Moderator: Are SOE debt and government-guaranteed debt usually included in a domestic debt restructuring exercise?

Patrick: The government has already said that guaranteed SOE debt will be included in the restructuring. It is still not clear at this stage if non-guaranteed debt will be included. But generally, SOE debt is not restructured if they are solvent on a standalone basis. But if they require government support, then there is a strong argument for including them, because otherwise the government will have to make transfers to service the debt stock. This is the case for most of Sri Lanka’s key SOEs. One interesting test case is the treatment for Sri Lankan Airlines, which required extensive government support in the past, but its debt has so far not been included in the restructuring. This could be a point of contention unless the company can show that its debt is serviceable on a standalone basis, without government support [Note: see here for an interesting analysis of the Sri Lanka Airlines debt in the FT].

Moderator: Assuming that there will be a domestic debt restructuring, where will you find greater NPV losses? Will it be on the short end of the yield curve, or will it be on the long end of the yield curve?

Patrick: I think the key difference will come down to T bills and bonds. As we discussed earlier, there is a chance that if domestic debt is restructured T bills are excluded from the process. But looking across the maturity curve for bonds, I think it is speculative to say which ones are going to experience more or less of an NPV burden. I think it is just too early in the process to tell.

Moderator: My next question is about exit yields. What is an exit yield, and how does it differ from a discount rate?

Patrick: For private creditors, the exit yield is the yield that the newly issued bonds initially trade at after the restructuring and will determine the initial hit they take in present value terms. For publicly traded bonds, the yield will fluctuate daily after the restructuring based on country-specific and external risk factors. Generally, yields are likely to fall if reform implementation is good and the restructuring restores sustainability. It might tend to rise if the restructuring is not viewed as credible, which is what we saw in Ecuador and Argentina's restructurings in 2020. Relatedly, the exit yield is the discount rate that is used to calculate the present value of relief offered by each group of creditors, and determines whether there is a comparability of treatment across participating creditors.

Moderator: What kind of exit yields do you anticipate for LK eurobonds?

Patrick: In normal times, we would use a range of c8% to 12% when we are modelling recovery values, depending on the credibility of the restructuring and reforms. But these are very far from normal times for emerging market debt. The yield on the high-yield emerging market sovereign debt index has risen above 12% from a post-Covid low of 6% just about a year ago. From 2010-19, the average exit yield was 7.5%. So, even for performing credits that are not in crisis right now, yields are already pushing double digits. Because of that, we have revised our exit yield range to 10% to 14%, with a 12% midpoint versus the 10% midpoint we would have used in the past.

Moderator: Should there be a single exit yield used for all creditors, or multiple ones depending on the risk environment in which each creditor is operating?

Patrick: This is a really contentious and important issue in debt restructuring. When the IMF calculates debt relief for creditors, they use an arbitrary 5% discount rate in their debt sustainability analysis. As I said before, market yields have now risen well into the double digits. So, a 5% discount rate is increasingly out of line with the rate that is relevant to private creditors in the open market. From this perspective, multiple exit yields – for instance, one for concessional bilateral debt, one for non-concessional bilateral debt, and one for commercial debt – may help to reflect the different risk and reward environments in which each type of creditor operates. On the other side of the coin, this might result in a perception that private creditors are providing less relief due to the use of a higher discount rate. So, this is likely to be a key point of dispute in the restructurings that are ongoing now, including Sri Lanka and Zambia. To resolve this over the longer term, I think it would be helpful to specify discount a rate in advance that will be used to calculate the present value of debt relief and to settle on a concrete and simple framework for how the present value of relief will be calculated.

But barring such a framework at this stage, I think it has to be hashed out at the negotiation table. If bondholders have their relief calculated in present value terms at a 5% discount rate, they are going to face much larger losses in present value terms based on a more appropriate market rate. I think this is going to be a big issue for them in the restructuring.

Moderator: You did a report a few days ago and issued a buy rating on LK eurobonds. What kind of recovery values and scenarios are you looking at? I would like to know your thesis for LK eurobonds.

Patrick: Sure, we did a recovery analysis back in March where we put a recovery value of 25-40 cents on the dollar at 12-14% exit yields for Sri Lankan eurobonds. As I mentioned before, I think the process is likely to be really long and difficult, for all the reasons we have articulated over the course of our discussion. But with bond prices dropping to 25 cents on the dollar this week, I think they are now attractive from a valuation perspective, with some potential upside. So, we did a quick back-of-the-envelope calculation where we figured out where Sri Lanka's bonds would break even at current prices. In our really bearish scenario, we did a 50% coupon cut, a 50% nominal haircut, and a five-year maturity extension. In this scenario, Sri Lankan eurobonds would break even at a 16% exit yield, which, as I said earlier, is really quite high even in the current interest rate environment. From that perspective, despite the fact that the restructuring is likely to be long and difficult, I think the bonds are now trading towards the lower end, if not below, recovery value, and are now attractive from a valuation perspective.

Moderator: Globally, how have banking systems been affected by domestic and external debt restructuring?

Patrick: In recent sovereign debt restructurings, the financial stability implications were assessed in advance using stress tests across banks and other non-bank financial institutions. The negative spillover effects can be managed through a combination of regulatory and crisis management measures which have included, for example, setting up a financial stability fund to provide solvency or liquidity support or, in the extreme, direct recapitalisation of banks by the public sector. Bank recapitalisation was only carried out in those cases where banks experienced significant capital shortfalls. But in some cases, banks were able to absorb those losses, including Barbados, Grenada, Saint Kitts & Nevis and Jamaica. So, it is not certain that if there is a domestic debt restructuring there will have to be a bank restructuring or recapitalisation. But really, the risk of that needs to be assessed in advance using stress tests. Then the costs and benefits of including that debt can be calculated in a more concrete way.

Moderator: Have you done an analysis of the Sri Lankan banking sector and the possible impact from both domestic and external debt restructuring?

Patrick: Yes. As I said earlier, there is very limited capacity for Sri Lankan banks to absorb losses currently. We calculated that with 5% nominal haircuts on domestic debt, Sri Lankan banks would require recapitalisation without regulatory forbearance. That rises to 14% if there are some regulatory forbearance measures put into place. But still, that points to a very limited loss absorption capacity. That means that if the domestic debt is restructured, then bank shareholders will likely have to be bailed in. I think external creditors will stop short of demanding that depositors are bailed into the process. But there will likely have to be some losses shared amongst the bank equity holders and potentially creditors due to the very limited buffers. By requiring the public sector to recapitalise banks, the benefits of including them in the restructuring could be washed away quickly. So, I think it will be required that the losses to be absorbed by the banks themselves.

Moderator: Let's assume, hypothetically, one of the bilateral creditors disagrees to reach a consensus with the rest of the creditors. Is that a possible scenario? In such a scenario, what are the options we can look at?

Patrick: None of the three major creditor groups, (1) Paris Club bilateral creditors, (2) Non-Paris Club bilaterals, or (3) commercial external creditors, want to agree to a restructuring deal until they know for certain that the others will accept comparable terms. If any creditor is reluctant to negotiate or does not disclose the terms of its debt relief, then the entire process essentially grinds to a halt.

In the past most external debt was extended by the Paris Club, but now, with the rise of eurobond issuances and other external creditors – like China, for example – that is no longer the case. This is part of the reason we are seeing much more complicated restructurings currently. In Sri Lanka's case, this is particularly an issue. Around 60% of its bilateral debt stock is owed to Non-Paris Club members, mostly China (which is c40%), and India (which is another 15%). Japan is the other major bilateral creditor.

We have seen delays in the recent Common Framework restructuring in Zambia and Ethiopia due to the reluctance of China to sit down at the negotiation table. So, in this context, initial statements of disapproval by various Chinese officials regarding Sri Lanka's decision to default are quite worrying. Ultimately, the process cannot move forward until all creditors – especially China, being the largest one – are willing to sit down at the table and negotiate.

Moderator: How long does it take for a country’s sovereign rating to get upgraded post-restructuring?

Patrick: The upgrade tends to happen immediately after the restructuring is finalised. The credit ratings typically will be upgraded once the process is complete from default/selective default to somewhere in the single B range. That said, it will take some time for ratings to go back to pre-crisis levels, potentially a few years even if the restructuring goes well. But we should at least see an immediate upgrade into that single B range once the restructuring is completed.

Moderator: How long does it take for a country’s sovereign rating to go to the pre-restructuring level? T+3 years? T+4 years?

Patrick: I think it depends on how successful the reform programme is. If Sri Lanka hits all its targets under the IMF programme, and the restructuring is successful in restoring debt sustainability, it could be two or three years. In Sri Lanka's case, even before the crisis, the country was trading in the single B range. So, it could reach that pre-crisis level more quickly than others based on the pre-crisis rating being so low in the first place.

Moderator: Okay. How long would it take for the sovereign to regain market access post-restructuring?

Patrick: One of the key arguments against a sovereign default is that it locks countries out of the market for many years. But in practice, we find that sovereigns typically regain market access within about two to three years after a debt restructuring. However, there could be a default premium associated with that. Basically, that means they will have to issue at rates higher than what they could have issued at before the default. The literature on this is mixed, though. Some studies have shown that it is a very small and short-lived premium, while other studies have shown that countries have to pay quite a bit more to issue debt after the restructuring. But generally speaking, Sri Lanka should be able to regain market access by the end of the IMF programme, which is about four years, if not sooner.

Moderator: On average, what kind of default premiums are we looking at? What has been the recent global experience?

Patrick: As I said, this is a controversial topic in the literature. Some studies say there is no premium at all, while an IMF study in 2015 showed a 400bps premium immediately post-default and a premium of 200bps after five years. This suggests that the premium could be quite large. Obviously, that is a wide range from zero to 400, but ultimately there is likely to be a bit of a premium, and it is likely to last a few years. But if Sri Lanka wants to re-enter the market within two or three years, and if they are successful in their reform programme, then they should be able to do so.

Moderator: Hypothetically, if there is a domestic debt restructuring, how will it impact the central bank's ability to raise money from the domestic market? What has been the global experience?

Patrick: The resumption of domestic market access is effectively immediate because it is a much more captive investor base. That said, real long-term domestic bond yields tend to rise sharply around a restructuring and it takes several years to revert to their pre-crisis levels. So, while Sri Lanka is likely to be able to continue issuing in the domestic market after the restructuring, it is likely going to be at higher real yields for quite some time.

Moderator: As per the IMF’s debt sustainability analysis on Sri Lanka, they are looking at a primary balance of +2.3% by 2025, from a projection of a negative 4% in 2022. Do you find this too very optimistic? What is your view on this?

Patrick: An adjustment of this magnitude is definitely very ambitious. It would be in the top decile of all IMF-supported programmes for low-income countries since 1990. From Sri Lanka’s perspective, the maximum primary surplus they have been able to attain since 1990 was a +0.6% of GDP, which occurred in 2018 before the tax cuts. The average balance in the past decade (pre-Covid) was a 1% deficit. So, no matter what metric you are looking at, a 6.3% of GDP adjustment over the course of three years and a +2.3% primary surplus is quite an aggressive adjustment. That said, there is a lot of low-hanging fruit. Sri Lanka’s revenue was 13.5% - 14.0% of GDP in the three years leading up to the 2019 tax cuts, and this compares to just 8.3% of GDP last year and an 8.8% target in 2022. So, there is clearly a lot of room to boost revenue. 2019 tax cuts, according to the President, cost an estimated 3.5% of GDP. The government has already reversed many of these tax cuts. That said, to get revenue back to the pre-tax cut levels, they are likely going to have to go even further with other new revenue measures because of the erosion of the tax base.

There is also some low-hanging fruit on the expenditure side. Capex was c4.5% of GDP over the past two years. That can be reduced at the margins. And subsidies have also been rather costly at c4% of GDP over the past two years. So, it is definitely going to be a difficult adjustment. But at the same time, I think there is a lot of low-hanging fruit that Sri Lanka can seize on to achieve those targets.

Moderator: If all goes well, do you expect a significant increase in foreign portfolio inflows to Sri Lanka post-debt restructuring?

Patrick: I cannot speak for equity flows, but on the fixed income side of things I think it is going to take some time. Real yields, which are sharply negative at the moment, will have to go into significantly positive territory. Before the restructuring, there was no significant participation in Sri Lanka’s domestic debt market by non-residents. We have seen an uptick in portfolio inflows over the past couple of weeks to cUS$60mn, which could indicate that some foreign investors are speculating on the possible exclusion of domestic debt from the restructuring. But, of course, it also depends on external conditions. Portfolio inflows are unlikely in the current environment of dollar strength and tightening global financial conditions. So, there needs to be a number of things that go right, both on the domestic and external front, before fixed income investors start looking at Sri Lanka as an attractive investment destination again.