Ukraine seemingly took a step closer to a sovereign bond default after state-owned Naftogaz announced a consent solicitation to defer eurobond payments for two years on 11 July. Nafto's decision may send an important signal about the government's willingness and ability to pay its own foreign bonds ahead of principal payment on 1 September, its first since Russia's invasion in February. The thinking is that if Naftogaz isn't paying, by extension, the government won't pay either.
Ukraine's government has hitherto remained current on debt service payments on its foreign bonds despite the war (as have other government-related entities including, until now, Nafto) although, even before Nafto's decision, markets began to factor in an increasing probability of non-payment, and investors are looking ahead at what default and restructuring look like. There is still enormous uncertainty about this (we don't even know what a map of Ukraine will look like), but we set out some initial thoughts.
We resist being too precise at this stage, calculating expected value based on scenario-weighted recovery values, given the uncertainty, but even if we weighted our three political scenarios (Ukraine wins, Russia wins, stalemate) 1/3, 1/3, 1/3 (ie we don't know), it is conceivable to derive recovery values above current prices; and, with prices so low, below US$20s beyond the 23s, downside is limited. However, recovery may also take time (although bondholders may be compensated to some extent by PDI if Ukraine defaults). Moreover, although we guestimate public debt has risen to 130% of GDP, implying debt reduction of the order 60%, there is considerable uncertainty about the amount of debt reduction required, and it could be more (or less).
We assign a Hold to the USD bonds at this stage (excluding the 22s, which we think will go into default), with UKRAIN '32s indicated at US$17.9 (mid-price basis) as of cob 14 July on Bloomberg.
Naftogaz's consent solicitation
Naftogaz announced its intention to seek to defer principal on the upcoming '22s maturity (due 19 July) for two years until 2024 and also defer interest until then. The principal payment amounts to US$335mn. The company is also looking to defer interest on two other bonds, the '24s (with a coupon also due on 19 July) and the '26s (due in November) until 2024 as well.
The consent solicitation, which also seeks to waive any potential event of default and certain covenants, requires approval through Extraordinary Resolutions in respect of each bond with the support of more than 75% of bondholders. The voting deadline is 21 July, with results due on 26 July. We don't know if bondholders will support the decision, and there might be some frustration after the company seemingly indicated previously that it was going to pay, according to reports, although it could argue it was forced into it by the government. If it isn't approved, the company could seek to sweeten the terms, or it could default anyway.
The market reaction suggests Nafto's announcement came as a bit of a surprise (with the '22s -7pts on the day, -35pts this week and -55pts from a week ago, now indicated at US$22.7 (mid price basis) as of cob 14 July on Bloomberg) although it shouldn't have been after the company said on 4 July that it was evaluating its liquidity requirements.
The decision may cast further doubt on Kyiv's ability (if not willingness) to pay amid talk that has emerged in recent weeks that the government is considering restructuring options, perhaps under pressure from donors and in the face of the sharp deterioration in its public finances.
But restructuring talk is despite the government's stated policy in the immediate aftermath of Russia's invasion to remain current and the demonstrated track record of making payments since then, including most recently state-owned Ukrainian Railways making its own bond payments this week and City of Kiev payments last month – albeit for much smaller amounts than is now due on Nafto bonds.
Ukraine Railways announced that it will pay both the coupons due this week at the same time – interest on the '24s due on 9 July and interest on the '26s due on 15 July, for a total amount of US$36mn. However, any positive demonstration effect from Ukrainian Railways will have been undone by Nafto's decision.
In all, this track record has seen Ukraine (and its state-owned enterprises) make some US$291mn in payments from the beginning of May until now, although this is mainly small interest payments.
The government of course also faces a US$912mn principal payment of its own on 1 September (on the UKRAIN '22s), where the market-implied probability of default (POD) has risen in recent weeks (interest on other bonds, too, takes total debt service due to US$1.2bn) – before then, there is also a small interest payment on 1 August on the 8.994% '24s, which we expect to get paid (not worth triggering default for). Until late June, the '22s were indicated at cUS$70, suggesting a POD of c50:50 (assuming a 40 cents recovery), while the rest of the curve was in the 30s (ie the market was expecting a default but not yet), prompting interest in curve trades among some investors.
However, restructuring talk had already seen UKRAIN '22s sink to below US$50 last week and they have fallen further this week, with the Nafto decision seen as increasing the likelihood of default on the UKRAIN '22s. The bonds were indicated at a low of US$37 as of cob 13 July on Bloomberg, although they rose 4pts to US$41 by close 14 July after the IMF said it expected Ukraine to continue servicing its debt. However, we think the market may have got ahead of itself here, as those comments were made in the context of the IMF's regular fortnightly press briefing; we'd expect the Fund to say that in that context.
Meanwhile, the longer bonds (UKRAIN '32s) are now indicated at cUS$18 as of cob 14 July on Bloomberg (mid-price basis).
But this may not be a surprise and, of course, the longer bonds were already priced for default anyway; indeed, many investors are surprised that Ukraine has remained current for so long under such exceptional circumstances.
We understand the government's initial wish to remain current on its bonds following Russia's invasion, so as to protect its reputation in the market as a recurring issuer and preserve market access (as it looks to commercial lenders to fund part of its massive post-war reconstruction costs, which the government recently put at US$750bn) – especially if the war proved short-lived and Russia backed down. However, we remarked that the longer the war continued, this position may not be tenable or even desirable amid more pressing needs.
Four months on, with costs mounting, public finances under pressure, delays to donor funding and with no certainty about how long the war will last (or what Ukraine will look like afterwards), we think bondholders would surely forgive a sovereign default, and behave in a cooperative and friendly way in an eventual workout given the exceptional circumstances. And, in any case, evidence tends to suggest that market access is restored surprisingly quickly, sooner than one might think, so Kyiv shouldn't be overly worried that it would never be able to return to the market.
Moreover, it might only be a matter of time – if it hasn't already been a topic of conversation – before IFIs and international partners question why donor money is being used to pay bondholders. This might be especially the case now amid recent reports that funding needs have increased, reportedly nearly doubling from US$5bn a month to US$9bn, while the EU's own US$9bn financial package announced in May has been subject to delays. The EU has, however, approved a further EUR1bn in macro-financial assistance earlier this month and the US has just provided a US$1.7bn grant.
So far, Ukraine has received US$12.7bn in external funding (from IFIs and bilaterals) out of total funding since the war of US$24.5bn, according to the Ministry of Finance (as of 12 July). That's c52%. The rest, and therefore a significant slice, has come from central bank financing (31%, the single biggest source) and domestic government bond issuance (17%). However, monetary financing is not sustainable.
In addition, notwithstanding the prospect of using confiscated Russian assets (which raises legal issues), with the official sector likely to bear most of the burden in terms of reconstruction costs (ie providing new money and mainly in grant form or highly concessional terms), it might argue that burden sharing from bondholders should involve debt relief; a similar situation to what we saw in the 2005 Iraq restructuring.
Meanwhile, restructuring could have also come up in programme discussions with the IMF in the context of an updated debt sustainability analysis (although any DSA at this stage would surely be subject to enormous uncertainty). The IMF noted in the DSA at the time of the request for emergency RFI support (March 2022) that debt was sustainable on a forward-looking basis, conditional on the quick de-escalation of the conflict, adequate and upfront financing from multilateral and official bilateral financing, and strong policy implementation. However, much has happened since then.
Stand still...and wait
The 1 September payment may therefore provide a deadline for the government to revise its debt management strategy. It could pay, and buy a bit more time (and, arguably, the incentives to pay the government bonds are different to Nafto). It could also be argued, although less convincingly in our view, that paying just US$1bn is a small price to pay to retain investor confidence. But, equally, there seems little point in delaying the inevitable, and wasting valuable resources.
But it would also seem premature to talk of restructuring modalities in any great detail in the midst of war, and with no sign of it ending, when we don't even know what the map of Ukraine will look like. It would be putting the cart before the horse somewhat.
Rather, as we have said before, organising a temporary debt standstill, similar to the debt service suspension initiative (DSSI), the G20's post-pandemic response to alleviate the debt service burden of low-income countries, may be an appropriate response at this stage. This could be done either formally via consent solicitation (like Ecuador did in 2020), or informally by gentlemen's agreement (ie bondholders don't elect to declare default and accelerate, or seek to cross-default, rather they just tacitly accept default and remain patient). The government, bondholders, IMF and IFIs can then figure out what a restructuring looks like later.
One problem with a standstill, however, is that it is hard to know how long it should last.
Although we think it is still too early to talk about recovery prospects with any certainty, we set out some initial thoughts here.
Ukraine sovereign bonds (outside the shortest tenors) are pricing in something like an 80% haircut (eg for the 32s, at a 12% exit yield). This is HIPC-style debt reduction, or Iraq-style (80% debt reduction on official debt and 90% on commercial debt).
This sounds excessive, but it is difficult to be sure about anything. It is not clear what the size of Ukraine's economy will be (pre-war nominal GDP was projected at cUS$200bn this year) if it loses significant territory under Russian occupation – the government says it has lost 20% of its territory. And there will be huge uncertainty about its economic structure and prospects; its productive capacity, trend growth, revenue base, export potential and import needs – especially if it loses the industrial heartland of Donbas and the ports of Mariupol, and even Odessa, which would leave Ukraine a much smaller land-locked country. However, there will be huge international goodwill (and financing) to support a free and independent post-war Ukraine and there is an argument that the Ukraine that emerges will be a stronger, more united country with a stronger identify (and less Russian interference and less oligarchic vested interests), and – probably – moving towards EU membership, which should be positive catalysts for post-war economic prospects.
But it is clear that public debt/GDP will have increased sharply (from c50% of GDP pre-war, according to the Ministry of Finance), signalling the need for perhaps sizeable debt reduction – in nominal or PV terms. This is due in part to higher debt (due to higher financing needs, although much of the external finance is in grant and concessional form) while the denominator will also be lower, with lower GDP (through economic damage and loss of territory), as well as through exchange rate effects (currency devaluation). The IMF RFI projected debt to increase to 60% of GDP but this looks like an underestimate.
For instance, simplistically, we assume debt/GDP has increased to 130%, with the numerator increasing by 27% (based on post-war budget financing received), while GDP may have fallen by 50% (according to government reports).
However, even if debt/GDP has increased, that should not determine haircuts today, rather that should be based on the longer-term projections for Ukraine's economy and its recovery. Still, that would suggest something like 60% debt reduction to return debt/GDP to its pre-war level, although not all the burden should fall on bondholders (assuming it did, for illustrative purposes, that would imply a PV of cUS$30 on the '32s).
Much will also depend on how the war evolves, and when and if it comes to an end, and what Ukraine looks like when it does.
For illustrative purposes, and to over-simplify, we identify three scenarios, although absent an outright victory either way (which seems unlikely) – either Russia pulls out of Ukraine or Russia take Kyiv – we think the outcome (end of the war) would most likely have to be agreed through a ceasefire and peace settlement, with whoever has the upper hand at the time in the driving seat for dictating terms, while economic damage to the country (in terms of physical and human capital) may increase as the war continues (even if some limited rebuilding is possible outside the main conflict areas).
Ukraine 'wins' – with territory returning to the pre-invasion situation, if not pre-Crimea annexation. This might minimise the losses to potential GDP and reconstruction costs. Might involve soft restructuring terms with long grace periods (and/or low coupons), but without the need for significant nominal haircuts, to defer payments (for a few, if not several, years) until repayment capacity has improved sufficiently.
Russia 'wins' – Russian occupation effectively partitions Ukraine, although the outcome would depend on what territory Russia held at the time. It would seem unlikely that Russia would willingly give up most of the territory it has gained since the invasion, especially in the Donbas and Mariupol, but it may give up some in return for concessions from the west (such as sanctions relief). We assume Kyiv would have no responsibility for Russian-occupied territory. The more territory Russia occupies, the lower recovery values would appear to be on what part of Ukraine remains (and perhaps a stronger moral argument to provide more generous relief to Ukraine) although perversely, perhaps the lower reconstruction costs would be too.
War of attrition, with a military stalemate (a long frozen conflict) – at some point, it may be deemed necessary to organise a restructuring, although, without a decisive outcome, agreeing restructuring modalities in the shadow of war could be tricky. This might suggest erring on the side of caution and aiming for a deeper restructuring than otherwise to build in some sensitivity to shocks.
We resist being too precise at this stage, calculating expected value based on scenario-weighted recovery values, given the uncertainty, but even if we weighted these scenarios 1/3, 1/3, 1/3 (ie we don't know), it is conceivable to derive recovery values above current prices; and with prices so low, below US$20s beyond the 23s, downside is limited. However, recovery may also take time (although bondholders may be compensated to some extent by PDI if Ukraine defaults). Moreover, although we guestimate public debt has risen to 130% of GDP, implying debt reduction of the order 60%, there is considerable uncertainty about the amount of debt reduction required, and it could be more (or less).
We assign a Hold to the USD bonds at this stage (excluding the 22s, which we think will go into default), after withdrawing our recommendation in the immediate aftermath of the war, with UKRAIN '32s indicated at US$17.9 (mid-price basis) as of cob 14 July on Bloomberg.
We leave the GDP warrants aside for the moment and aim to look at them in more detail another time.