Brad Setser is a senior fellow at the Council on Foreign Relations and a former Treasury Department official. Theo Maret is a research analyst at Global Sovereign Advisory and writes a sovereign debt newsletter.
When Zambia announced an agreement with bondholders to restructure its three outstanding Eurobonds, many thought the G20’s Common Framework would finally be able to notch its first major success. Whoops.
Alas, the deal was rejected several times by the official sector and the way forward is now unclear. Bondholders are irate. At the heart of the intra-creditor skirmish lies the thorny question of what constitutes comparable treatment for different creditors.
Comparability has been the longstanding norm linking the treatments of official and commercial claims. But the assessment is now complicated by a fragmented creditor landscape, limited transparency about the stock of official claims and a lack of trust in existing methodologies, which have grown rusty in the absence of regular use prior to the pandemic. Let’s dive in.
Comparability 101
Any bond restructuring for Zambia must pass two tests: it must meet the IMF program targets, and it must be judged to be comparable with the complicated deal Zambia reached with its official creditors back in late June.
IMF targets include indicators derived from the Fund’s debt sustainability analysis — for Zambia, external-debt-to-exports and external-debt-service-to-revenue ratios — as well as the closing of balance-of-payments financing gaps for each year of the IMF program.
Comparability of treatment is the idea that the terms of the private bank and bond restructurings should be “comparable” to the terms of the restructuring of official bilateral creditors, ie loans made by other governments and government-backed export credit agencies.
To assess comparability, the Paris Club uses three different formulas: nominal debt service relief over the IMF program period, extension of the duration of the claims, and reduction of the debt stock in present-value terms (see this World Bank note for more details). Official creditors then make a judgment based on the three comparisons.
The inclusion of China among official creditors has generated pressure to firm up the comparability requirement. China isn’t keen to subsidise private bondholders which it considers have gotten too sweet deals in the past (private creditors notably did not participate in the 2020 G-20 Debt Service Suspension Initiative). At the same time, traditional bilateral lenders aren’t keen to subsidise Chinese lending that is sometimes perceived as . . . reckless.
Making sense of what happened in Zambia
Zambia’s debt stock is complex, with a mix of official creditors — the largest being the Export-Import Bank of China with over $4bn owed — another $3.85bn in Eurobond claims, and about $3.5bn owed to commercial banks, including large Chinese state banks. There’s also $2.7bn in non-resident holdings of local currency bonds, excluded from the restructuring but counted as external debt by the IMF.
Virtually every substantive issue possible in a restructuring is thus on the agenda (zoomable version).
All the theoretical considerations about IMF targets and comparability became very real for Zambia when the IMF and official creditors — notably China — in November rejected a bond deal that provided a roughly equal NPV haircut to the official sector deal, but differed on other dimensions.
The IMF’s objections, luckily, were narrow — Zambia and bondholders agreed to make the adjustments needed in an updated deal. Yet even with the adjustment, the Official Creditor Committee (OCC) did not bless the deal.
Bondholders have provided a table showing how their proposed deal scores on the three comparability criteria. The OCC did not push back on the below numbers, so we’ll take them at face value (pun intended).
NB, it’s been reported that a third version of the deal was presented by the country to official creditors, and rejected again, but no details have publicly emerged, and we understand it is close to these last public figures (zoomable version).
The table shows that bondholders have a strong preference for short-term cash flows, and thus the debate becomes how much additional present-value reduction is needed to compensate for more upfront cash — the 1 percentage point embedded in the base scenario is apparently not enough for the Paris Club and China.
This stalemate illustrates the main issue with the current implementation of comparability: it’s ultimately a subjective judgment, making it impossible to know in advance how concessions on one dimension need to balance with greater effort on others.
No one knows for sure how to trade NPV for upfront cash: per Zambia’s statement there is not even consensus among members of the Official Creditor Committee on this critical issue, arguably the result of having new players like China at the table. Bondholders understandably aren’t keen on continuing a FAFO “try and see” back-and-forth with the official sector, arguing that the official sector is intervening in their own negotiations with the country.
To add additional intrigue, Bloomberg reported that China obtained approval from the authorities in Beijing to sign off on the June deal based on indicative terms for bondholders that were substantially harsher — 10 percentage points of additional present-value reduction — thinking these terms were final. The bondholders weren’t told about this until later. Other bilateral creditors have suggested a roughly 5 percentage points difference in present-value reduction between the official creditors and bondholders would work.
What is the price of a haircut?
One argument raised by bondholders is that their principal haircut should be taken into account when assessing comparability, in effect adding a fourth substantive indicator.
However, the argument only really gains political merit when the face value of the new debt instruments is being reduced below its level at the time the country first got into trouble. Zambia’s restructuring process has been exceptionally slow. Accumulated past due interest (PDI) has added enough to the bondholders’ legal claim that the face value of the new bonds is set to be higher than that of the old ones, even after the proposed haircut.
Bondholders are not alone: official creditors have also accumulated past due interest, yet they have not properly disclosed the size of their claims pre- and post-restructuring — it would help if they were more transparent.
To be sure, PDI is a contractual right, and bondholders weren’t responsible for the lag between Zambia’s default and the start of real negotiations — they would rather have had performing bonds than a growing legal claim. Proposals to scrap PDI altogether go too far; a better approach would respect contractual claims without allowing the rise in claims to pull up estimates of what the country actually can pay.
The classification of Chinese claims comes back to bite
Zambia faces another complexity. The contemplated bond treatment would apparently not leave enough cash flows within the IMF envelope if other commercial creditors were to take the same deal. As a Paris Club official told Reuters:
If bondholders’ debt relief fell short of expectations, that also raised questions about the effort that would be needed from other private creditors like banks and some Chinese institutions.
IMF thresholds indeed are a zero-sum game: commercial creditors standing last in line might be required to do “more” than official creditors or bondholders to fill the restructuring envelope.
Historically this was not much of a problem, but in Zambia’s case, China concluded that only China Ex-Im was an official creditor: in June 2023 $1.7bn in claims backed by China’s export credit agency — originally counted as official — were reclassified as commercial.
In hindsight this policy choice created a bit of a mess: all the Chinese “commercial” claims are owed to state-owned entities, so it isn’t clear if China’s representatives on the OCC are protecting the interest of all Chinese state creditors or simply these of China’s designated official lenders.
China’s decision is even backfiring, as its “commercial” banks cannot extract better terms than the Paris Club since comparability is enforced, while bondholders moving first can obtain better terms than Chinese banks without being subject to comparability with other commercial claims.
The obvious fix for this procedural nightmare would be for China to put all the claims of entities controlled by the state in the official bucket and empower a single negotiator to represent the interest of all such entities — or even better, transfer all the distressed loans of its policy banks to a bad bank.
What’s next?
Bottom line, something will have to give.
The official creditors should help sort out the immediate mess by clarifying how much present-value reduction is needed in exchange for bondholders getting the lion’s share of available cash upfront. Then, tweaks in the bond deal potentially could help get something close to the last agreement over the finish line.
Creditors are taking advantage of the large amount of dollars that the IMF program leaves available in coming years: it allows Zambia to pay about $1bn a year in external debt service in 2024 and 2025, while net reserves are below $2bn and expected to remain under $2.5bn. This high debt servicing in fact corresponds to a surge in official inflows in 2024 and 2025. This should raise questions about future program design: using an increase in preferred debt (from the IMF and multilateral development banks) to allow private creditors to exit generates additional risks to both the Fund and the borrower down the road.
But taking a step back, beyond jerry-rigged fixes for Zambia, the time is ripe for a rethink of how best to define comparability.
There’s a trade-off between the need for clarity and simplicity — which a unique mathematical formula would provide — and the inherent political aspect of comparability, which has always reflected the specific constraints faced by major bilateral creditors.
Right now though, with a broad sense of distrust across the board, there’s a clear need to shift towards the prioritisation of upfront clarity. In Zambia, official creditors agreed to a deal based on the assumption that other creditor groups would accept a significantly higher present-value reduction, and the bondholder deal would not work if other commercial creditors asked for a similar amount of upfront cash.
Clarity should start with transparency about the overall scale of the effort needed and the amount of cash available for all creditor groups that are part of the restructuring.
The core data needed for the calculations — NPV of claims, interest and amortisation by creditor group during the program, available foreign exchange for external debt service — should be in the actual financing tables of IMF staff reports (the best data currently comes from an outdated investor presentation). This would make it immediately clear if a creditor group was doing a deal that implied an asymmetric effort by other creditors.
Some adjustments to the mathematical formulas might then be appropriate, dropping the duration extension (which is redundant) to focus on present-value reduction and the allocation of short-term cash flows. To limit the distorting effect of PDI and incentivise faster deals, the present-value reduction used for comparability could be calculated using either the original face value or the size of the claim at the time of an IMF staff level agreement as the denominator.
Apportioning the available cash on the calculated stock, not claims due in the program period, would also makes sense in cases of deep distress. Trades that offer deeper haircuts for more front-loaded cash are in fact often good all around, but they need to be assessed against a clear baseline. Symmetric treatment of the stock inside the perimeter of the restructuring is an obvious choice.
Such an approach would create a safe space for quick deals. Any creditor group that agrees to a deal that passes a basic test for maturity extension and does equiproportional (symmetric) NPV debt reduction for an equal share of the available cash flow would not need to wait for other creditor groups to approve the deal. They would of course want a chance to reopen their deal if the issuer and the IMF subsequently agreed to increase the size of the pie (through updates of program parameters, or with state contingent instruments).
Of course, this new norm would also place an even bigger premium on the IMF’s ability to rapidly set out parameters ensuring a reasonable balance of effort between debt relief and fiscal adjustment.
Squabbles over who gets the biggest slice of the pie are less important than getting the size or the flavour of the pie right: the real debate should be over the level of debt that the issuer can support, with and without state contingent features, not which creditor can score a better deal than others.
Source: Economy - ft.com
