Sri Lanka's IMF program: connecting the dots
Sri Lanka reached a Staff-Level Agreement with the IMF in September on a USD 2.9 billion arrangement under the Extended Fund Facility. The next hurdle towards Executive Board approval and disbursements is the requirement for the country to seek financing assurances from creditors, which might prove trickier than anticipated.
In this post I try to disentangle several intertwined challenges, first by giving a brief overview of the IMF financing assurances and arrears policies, before looking at the institutional setting in which the negotiations could take place in Sri Lanka. Then I discuss briefly specific aspects related to the participation of China in the negotiations, and conclude with a broader perspective on the precedent that Sri Lanka could set for the international financial architecture.
A brief reminder of IMF financing assurances and arrears policies
The IMF states that “The financing assurances policy requires Fund-supported programs to be fully financed”. Countries are required to obtain “firm commitments” to bridge the external financing gap for the next 12 months, e.g. with upcoming signed loan disbursements, and “good prospects” for the remainder of the program period. Good prospects then need to be gradually turned into firm commitments on a rolling basis through the program review process.
In addition to assurances to bridge financing gaps, countries with unsustainable debt will also be required to provide another type of assurances proving that debt sustainability will be restored. These come usually in the form of new non-debt-creating flows such as grants or debt restructuring and relief.
Once arrears have been incurred, as is the case for Sri Lanka which suspended most external debt repayments in April 2022, the IMF turns to its arrears policies which come in addition to the above assurances and are usually treated simultaneously, e.g. a debt restructuring that would clear arrears, restore debt sustainability and contribute to bridging any remaining financing gap by reducing debt service.
The IMF’s arrears policies are threefold:
The Lending into Arrears (LIA) policy for arrears to commercial creditors
The Lending into Official Arrears (LIOA) policy for arrears to bilateral official creditors
The Non Toleration Policy (NTP) for arrears owed to multilaterals
The level of assurance required for the IMF to approve a program in the presence of arrears hence depends on the type of creditor, or on the relative importance of the creditor in the country’s debt stock. First, arrears to multilaterals are a peculiar occurence and they fully prevent IMF financing apart from a very narrow set of situations.
For commercial creditors, the Fund can approve a program before a restructuring agreement has been reached as long as the authorities are making a good faith effort to reach said agreement with their commercial creditors. Good faith is assessed through a set of criteria described here (Box 1).
The harder part comes with bilateral creditors and the LIOA, for which there are three options (for the details, see again here, Box 2):
The country has a representative Paris Club Agreement
Bilateral consent has been provided for Fund financing despite the arrears
As a last resort, the country is making a good faith effort to reach an agreement with its creditors
It is worth noting that this last option of good faith was engineered mostly to deal with the situation of Ukraine and its bond owed to Russia in 2015, hence is very rarely an option considered in regular program negotiations.
For Sri Lanka the heavy-lifting in coming weeks – and the focus for the rest of this blog – hence lies in getting assurances from bilateral official creditors that they will commit to a debt treatment and/or new financing consistent with the parameters of the IMF program.
Sri Lanka in uncharted territories
In a presentation to creditors, Sri Lankan authorities announced that they would promote the creation of an ad hoc forum to coordinate all bilateral creditors, aiming to streamline the provision of financing assurances (see the slide below).
This proposal is interesting in several regards. First, 66% of Sri Lanka’s bilateral debt stock is owed to non-Paris Club members, mostly China (52%) and India (12%). Therefore, an agreement with the Paris Club alone could not be considered as representative by the IMF and deem away arrears to remaining bilateral creditors. Sri Lanka needs to find a way to get China (and probably India) on board.
Second, Sri Lanka is not eligible to the G20 Common Framework for Debt Treatment beyond the DSSI (CF for short), to which only 73 low-income countries are eligible. There have been repeated calls to broaden the scope of the CF to highly indebted middle-income countries hard-hit by the COVID-19 pandemic and subsequent energy and food price shocks, not least from the IMF itself or the US Treasury.
Since Sri Lanka is in effect trying to engineer its own CF process, it is worth taking stock of the current situation with CF cases when it comes to financing assurances.
The IMF discusses the CF in the latest update of its arrears policies (paragraph 53 onwards). The phrasing is a bit ambiguous: while the IMF signals openness to grant the CF a similar status to that of the Paris Club under the LIOA policy, it acknowledges the shortfalls of ongoing CF processes in Chad, Ethiopia and Zambia, and indicates more experience is needed for a change of the policies. At the same time, the IMF indicates the following:
Adequately representative CF treatments can only “deem away” arrears to non-participants when a Paris Club creditor participates (in such cases, both the non-Paris Club and Paris Club financing under the CF would count to determine representativeness, as is currently the case in Paris Club Plus treatments)
Since there is de facto always a Paris Club member in CF Creditor Committees – France even co-chairs the Zambia committee without having any claim on the country – it means that the CF will be able to benefit from some form of specific status under the LIOA policy.
Another important aspect is the level of commitment required for financing assurances within the CF process. Sri Lanka might be looking at Zambia in this regard, where the Creditor Committee published a statement that was considered by the IMF as sufficient financing assurances, despite not saying anything about the specifics of the upcoming debt treatment:
Consistent with their national laws and internal procedures, creditor committee members are committed to negotiate with the Republic of Zambia terms of a restructuring of their claims to be finalized in a Memorandum of Understanding (MoU), in accordance with the “Common Framework for Debt Treatments beyond the DSSI”.
This phrasing looks rather vague, and one could be surprised that the IMF approved the program without knowing for instance the flow relief that China could provide to Zambia over the program period, or whether it is ready to accept a haircut. However, Guillaume Chabert from the IMF noted at a recent conference that it was unprecedented to have a written commitment from China, and that it was the result of hours of negotiations. It might also be that the IMF got some informal feedback from China which was not made public but brought additional confidence – Jeromin Zettelmeyer described something similar for Suriname in a podcast.
An even more striking statistic: Guillaume Chabert also said that the Chad CF process and then Zambia were the two first occurences when China provided financing assurances for the IMF to approve a program.
Sri Lanka, not being eligible for the CF, will have to build on these recent developments and answer two major questions: how to get bilateral creditors (especially China) on board with the ad hoc platform, and what level of commitment to seek from them in this forum in order to reach the Board Approval for its IMF program – e.g. whether a statement similar to that of Zambia’s bilateral creditors would be sufficient.
Uncertainty surrounding the attitude of China and India
The provision by China of financing assurances outside of the CF hence remains untested, so it would be rather ambitious to draw definitive conclusions from the Zambia case. Some argued that China did take a lot time to provide assurances – 7 months between the Staff-Level Agreement in Zambia and the provision of financing assurances – because of a lack of experience, and that it would be faster for Sri Lanka. However other factors, not least geopolitical ones, could come into play and disrupt the Sri Lankan process.
An interesting precedent in this regard is also Suriname, where the IMF approved a program without clear financing assurances from China and India. The Fund writes in the initial Staff Report:
China and India have provided assurances, although less specific than those provided by the Paris Club creditors, that they intend to work with Suriname towards a debt restructuring that will restore sustainability. China has consented to Fund financing notwithstanding these arrears. India has requested more time to consider consenting to Fund financing notwithstanding these arrears.
And then in the first review:
Paris Club creditors have provided specific and credible financing assurances indicating that they will provide debt relief in line with program parameters. China and India continue to consent to the use of Fund resources despite Suriname running arrears on their official debt
Jeromin Zettelmeyer gave a more thorough explanation in a podcast, indicating that the IMF judged in Suriname that it would ultimately be in the interest of China and India to restructure their claims.
Whether the IMF could act similarly with regards to China and India in the case of Sri Lanka – not requiring a formal commitment to debt treatment – is highly uncertain, notably since political stakes are running high and pressure has been increasing vis-à-vis the involvement of China in debt restructurings.
An additionnal hurdle will be to define the scope of Chinese creditors to be considered as bilateral creditors. As Brad Setser pointed out, the IMF in Zambia seemed to opt for a broad scope for bilateral claims which might include banks like CDB or ICBC that China claimed were commercial creditors. This question might especially arise in Sri Lanka for so-called “indirect exposure” to China consisting of ECA-backed claims amounting to more than USD 3.2 billion, which are for now listed as bilateral claims in the creditor presentation (slide 16).
In order to reduce inter-creditor tensions and coordination problems going forward, it would be interesting to have a clearer view ex ante of the qualification of Chinese creditors. The IMF could engineer a similar classification methodology to the one it uses for multilaterals within the arrears policies, in the form of a binary decision tree assessing the treatment of this creditor in previous restructurings, the breadth of its membership, etc.
Some specific criteria to assess whether a bank is to be considered as a bilateral creditor could include financial ownership by the state or the presence of officials in the governance structure. In their “How China Lends” paper, Anna Gelpern and colleagues note that some loan contracts include the termination of diplomatic relations with China as an event of default: such clauses could be considered as supporting factors for the inclusion of loans in the bilateral debt stock, though obviously they are rarely available to the public or even IMF teams.
Finally, once the scope has been agreed for what constitutes the bilateral debt stock of Sri Lanka, there might be further internal tensions among Chinese creditors, as Brad Setser noted a few weeks ago when we discussed the Zambia program. The Chinese government might indeed have a say on how to allocate the debt relief among different entities and policy banks.
The ball remains in the IMF’s court
The IMF will be the final judge of what can be considered as sufficient financing assurances to approve the program in Sri Lanka and make the first disbursement. The financing assurances and arrears policies do provide a lot of caveats and wiggle room for judgement calls, which might turn the tide in either direction.
Sri Lanka will in any case create a precedent for sovereign debt restructurings after the COVID-19 pandemic in middle-income countries. For instance the attitude of the IMF and the application of its policies with regards to the ad hoc coordination platform will be very interesting to watch. Since a political push to significantly improve and expand the Common Framework in coming months seems unlikely at the G20, ad hoc processes will be the only choice for many countries seeking debt treatment, e.g. small middle-income island countries.
A glimmer of hope might come from the apparent willingness of the US – the main IMF shareholder with a de facto veto on all major decisions – to move forward swiftly on Sri Lanka, including outside of the Common Framework. It was best illustrated in the conclusion of Brent Neiman’s landmark speech on sovereign debt:
Sri Lanka urgently needs a debt restructuring and, unfortunately, is not eligible for the Common Framework. (…) We are in new territory, but the coming weeks offer a real opportunity for progress. (…) Swiftly concluding the first Common Framework case, and making clear progress on multilateral restructuring outside of the Framework, would be a big win not only for current and future debtors and their citizens, but also for all official creditors, whether traditional ones or new ones. We must build on recent experiences, apply lessons learned, and push ahead in these cases to prevent any depreciation of our global financial infrastructure.
What is at stake appears to be more than the fate of Sri Lanka, rather to some extent that of the international financial architecture for sovereign debt restructurings. Interesting times ahead.
I am grateful to Sophie Borra and Martin Kessler for valuable comments, and to Brad Setser for his insights and guesses on all things IMF Arrears Policies.
If you want to discuss this further, please reach out: firstname.lastname@example.org
You can also follow me on twitter @TheoMaret