Sovereign Analysis /

IMF/WB virtual Annual Meetings: 8 key themes

  • We summarise the discussions we had at the recent Annual Meetings

  • Our views are grouped into eight themes and specific country discussions. Our meetings covered some 16 countries in all

  • Global themes: borrowing space, monetary policy limits. Country highlights: Egypt, Ethiopia, Mozambique, Tajikistan

IMF/WB virtual Annual Meetings: 8 key themes
Stuart Culverhouse
Stuart Culverhouse

Chief Economist & Head of Fixed Income Research

Tellimer Research
25 October 2020
Published byTellimer Research

We attended the IMF/World Bank Annual Meetings (this edition virtual) over the week of 12-16 October, hosting a number of conference calls with IMF mission chiefs, resident representatives and officials from various country delegations – our meetings covered some 16 countries in all. We wish to thank all of our hosts for their time and for the constructive and open nature of the discussions, which were valuable to us and to the investors attending.

This report relates some of the ideas that came out of those meetings, and updates our views on the sovereigns that we cover. That said, the views we set out are our own – we make no changes to our recommendations in this note.

We group our views into eight themes and specific country discussions, as set out below:

  1. Space to borrow. The IMF is now urging governments with the ability to borrow at low rates to do so to alleviate the impact of the crisis, the corollary being there is no need to pursue fiscal consolidation too early and kill the recovery. This might be welcome news for some, but it may be less applicable to frontier or smaller EMs, where the rise in public debt burdens may already be above debt-carrying capacity and domestic revenue mobilisation will be the main challenge.

  2. Limits to monetary policy. Many EMs may have already eaten up most of the monetary policy space they had upon entering the crisis, and, while developed markets have been able to pursue unconventional monetary policy and quantitative easing, that is not an option for all. This could test monetary policy credibility in EMs and put more pressure on fiscal levers. 

  3. Positive surprises: Egypt and Ethiopia. Egypt is most notable, with its 2020 growth projection having been revised up from 2% to 3.5% and its broader outlook also displaying material improvement. Ethiopia’s outlook was not revised materially between April and October, but our meeting with the central bank struck an optimistic tone, with real, fiscal and external indicators surprising notably to the upside.

  4. Fiscal woes and elections: Ghana and Bolivia. Both countries have something in common – double-digit fiscal deficits this year, in an election year, and doubts about the commitment to, and speed of, fiscal consolidation, where they don’t have the luxury of space to borrow.

  5. Safe havens: Senegal and Cote d’Ivoire. Relative safe havens during Covid helped by stronger fundamentals coming into the crisis, which enabled them better to absorb the shock and allowed polices to respond appropriately.

  6. Mozambique versus Tajikistan. These two seemingly disparate countries share some common features, with massive investment projects. For both, the big challenge is getting from A to B (ie from here to project completion). But we think it is easier to see a path for Mozambique than Tajikistan.

  7. Ukraine: Programme wobbles. It is now well known that the IMF’s programme review is delayed as the government’s reform commitment has come into question. This is problematic in light of the country’s large funding needs.

  8. DSSI 2.0. This was a discussion topic in several of the calls we had on low-income developing countries. To no great surprise, the G20 confirmed during the Annual Meetings a six-month extension to the debt service suspension initiative (DSSI) until June 2021.


Economic outlook better in developed markets, unchanged in EMs, worse in LIDCs, but with some positive surprises

Since June, the global growth outlook has improved, with the IMF now forecasting a 4.4% contraction versus 5.2% in June. The upward revision results from better-than-anticipated Q2 growth outturns in developed markets (DMs), offset partially by lower H2 growth forecasts amid persistent social distancing and stalled reopenings.

However, the headline figure masks a divergence between DMs, emerging market and middle-income economies, and LIDCs. While the growth impact will still be greatest in DMs and smallest in LIDCs, the 2020 growth forecast has been revised down in EMs and LIDCs even as it has improved in DMs.

Within Sub-Saharan Africa (SSA) specifically, the IMF now projects real GDP to fall by 3% this year, its worst outcome on record for the region, although the impact is not homogenous and some countries are faring better (see here). Tourism-dependent economies and commodity exporters will likely see even bigger declines, however. A rebound is expected next year, but a return to 2019 levels is not expected to occur until 2022-24. That’s potentially five years lost. We think there are also risks around possible economic scarring and whether the crisis will lead to semi-permanent changes to traditional growth drivers in some countries (especially tourist destinations) – a potential issue for the Caribbean and other small island states too – while policy space is also more limited.

Similarly, the global budget deficit has been revised down from 13.9% to 12.7%, but this has been driven by improvements to DM outcomes totalling 2.2% of GDP between June in October versus marginal deteriorations of 0.1% of GDP across EMs and LIDCs.

2020 growth forecasts by WEO vintage
2020 budget deficit forecasts by WEO vintage

Despite a broad deterioration in the short-term economic outlook across EMs and LIDCs, there are several countries on which we held meetings that surprised to the upside. Egypt is most notable, with its 2020 growth projection having been revised up from 2% to 3.5% and its broader outlook also displaying material improvement (see here for a full summary of our Egypt IMF meeting). Other countries with notable improvements include Nigeria (+1.1%) and Ukraine (+0.5%).

While Ethiopia’s outlook was not revised materially between April and October, our meeting with the central bank (National Bank of Ethiopia – NBE) struck an optimistic tone, with real, fiscal and external indicators surprising notably to the upside (see here for a full summary of our NBE meeting). As such, we have become notably more sanguine on the economic outlook for both Ethiopia and Egypt on the basis of last week’s meetings.

GDP revision since latest forecast

No need for austerity if countries can borrow cheaply, but major divergence in EM/LIDCs

Another surprise has been a shifting narrative on austerity, with the IMF urging governments with the ability to borrow at low rates to do so to alleviate the impact of the crisis. The IMF’s Fiscal Monitor (see here) “emphasizes the importance of not pulling the plug of fiscal support too soon, in spite of the high levels of debt prevailing worldwide.”

On the one hand, public debt is expected to jump sharply from 83% of GDP in 2019 to 98.7% in 2020, with an increase of 20.2% of GDP in DMs, 9.6% in EMs and 5.5% in LIDCs. However, debt is expected to stabilise next year as higher deficits and lower growth are offset by the “low for long” interest rate environment, improving the r – g differential and informing the IMF’s relatively sanguine fiscal outlook.

Public debt forecasts

For countries with room to borrow, the IMF emphasizes the importance of investment in health and education and in digital and green infrastructure, with an entire chapter of the World Economic Outlook (WEO) dedicated to mitigating climate change.

However, here too there is significant divergence between DMs and EM/LIDCs. While low interest rates are a mitigating factor for DMs, the ratio of debt service to tax revenue is expected to exceed 20% in over half of LIDCs, leaving little room to borrow (and far lower fiscal multipliers due to resulting pressure on interest rates and private sectors being crowded out).

Within EMs and LIDCs there is also much divergence. In countries like Egypt, Ethiopia, Rwanda and Turkey, the IMF appears comfortable kicking deficit targets down the road due to existing fiscal space and/or a credible commitment to post-Covid consolidation. However, in countries that entered the crisis with elevated debt vulnerability and where deteriorating global risk sentiment could trigger a sudden stop in new lending (or failure to rollover existing debt), austerity remains an urgent priority.

Fiscal stimulus much larger in DMs than EMs and LIDCs

These factors have driven notable difference in fiscal responses across countries, with the amount of stimulus displaying a clear positive correlation with initial income level and negative correlation with initial sovereign spreads (we have previously written about this here).

Globally, Covid-related fiscal stimulus has totalled US$11.7tn (nearly 12% of GDP) through 11 September, roughly half of which consists of additional spending or foregone revenue and half of which consists of liquidity support including loans, guarantees, and capital injections by the public sector.

However, the vast majority of this has taken place in DMs, with stimulus measures totalling 20.2% of GDP vs 5.9% in EMs and 1.8% of in LIDCs. Furthermore, the composition is skewed towards off-budget measures in DMs, while in EMs it is skewed slightly towards on-budget spending and in LIDCs it is almost entirely on-budget.

Covid-related fiscal spending

However, here too there is significant heterogeneity among EM/LIDCs. In 16 countries (including Brazil, Oman, Mongolia, Kuwait, Fiji and Iraq, among others) the deficit is projected to widen by double digits relative to the 2019 outturn; in nine countries (including Egypt, Tanzania, Zimbabwe, Kenya, Cameroon and Ethiopia) it is projected to widen by 1% of GDP or less; and in nine others (including Sudan, Zambia and the DRC) it is actually projected to tighten.

EM/LIDCs have more ‘traditional’ firepower but less room for ‘unconventional’ policy

Fiscal stimulus has been complemented by US$7.5tn of balance sheet expansion. Asset purchases have been driven by DM central banks, with the four major banks (the Fed, ECB and Banks of Japan and England) purchasing on average 63% of new government debt issued since February. Meanwhile, among major EM central banks that have implemented asset purchases, this figure stands at only 30%.

Central bank purchases of government debt

A greater reliance by DMs on ‘unconventional’ measures like asset purchases and QE has been driven namely by low policy rates pre-Covid, leaving little room for rate cuts. Meanwhile, most major EMs had ample room to cut rates, with countries like South Africa (see here for our discussion with South Africa Reserve Bank Deputy Governor Cassim) using that space to aggressively cut rates by 275bps since the crisis began.

Average central bank policy rates

However, elevated external vulnerability may limit the extent to which monetary policy can be used to further ease the impact of the crisis, with South Africa in a ‘wait and see’ mode and Turkey hiking rates. Meanwhile, other countries such as Ethiopia (see here) are in the process of modernising their monetary policy framework and have little room to implement stimulus of any kind at the moment.

Despite more “traditional” policy space in EMs versus DMs, external funding and/or credibility gaps likely limit the extent to which this space can be used (see here for our previous analysis on this topic). This puts the onus firmly on fiscal policy to stimulate the economy, which is problematic in countries like South Africa with elevated fiscal vulnerability.

Country-specific discussions

Egypt and Ethiopia: Positive surprises

The two main positive surprises to come out of the meetings last week were Egypt and Ethiopia, where the fundamental outlooks appear to have notably improved.

In Egypt, growth has held up better than expected, with the government’s preliminary estimate for FY 19/20 (ending June) at 3.6%, higher than the IMF projection in June of 2%. The IMF is also considering a revision of its 20/21 projection from 2% to 2.5-3%. The current account deficit was also stronger than expected in 19/20, reaching 3.1% of GDP as remittances were surprisingly durable and tourism performed modestly better than expected. These trends were broadly offset by higher-than-expected imports, in line with the higher-than-expected growth.

On the fiscal policy front, Egypt achieved remarkable stability at the height of the Covid crisis, with a primary surplus of 1.8% of GDP in the year ending June. However, the government is open to further stimulus amid heightened demand for social assistance and support to the health and tourism sectors, so a deviation from the government’s consolidation path will likely be warranted (in line with the general tone of the meetings that countries with room to borrow should do so to offset the impact of the crisis). The primary surplus floor under the IMF program is 0.5% of GDP, so there is space for c1.3% of GDP worth of stimulus. Post-Covid, the 2% primary surplus target is certainly within reach if growth recovers in line with expectations, but may take longer to reach if the recovery is weak.

Although the risk of capital outflows has increased as the Egyptian carry trade has again become crowded, reserves are healthy enough to prevent a disorderly adjustment, the IMF views EGP as being fairly valued, and low spreads and ample demand for recent US$ issuances show that investors seem to believe in the fundamental story as well.

Meanwhile, Ethiopia also looks set to emerge from the crisis in relatively good shape, and is projected to record growth of nearly 2% in 2020 (a level surpassed only by Guyana, South Sudan, Bangladesh and Egypt). This has been supported by stronger-than-anticipated exports and remittances, with exports actually rising 12% yoy in H1 and remittances contracting less than anticipated.

On the fiscal front, the FY 19/20 deficit is estimated to have been broadly unchanged at 2.5-2.6% of GDP, with taxes actually rising 17.6% on the year despite the Covid-induced slowdown and bucking the recent trend of revenue under-collection. In addition, public sector external debt actually declined from 28.2% in the previous year to 26.6% of GDP. Despite increased social spending, the deficit is likely to remain broadly unchanged this year too, owing to conservative projections.

Although the macro outlook has generally surprised on the upside post-Covid, past growth has been driven mainly by public investment and there is a need to shift to a more private sector-driven growth model. To this end, the government has embarked on a three-pronged reform program focused on macro, structural and sector-specific reforms. The NBE’s reform plans are particularly ambitious, including modernisation of the monetary policy framework, financial sector deepening and liberalisation, and movement towards FX flexibility. If the government sticks to its reform plans the macro outlook is sure to improve, although it remains to be seen if reality will live up to expectations.

We retain our Buy recommendation on Egyptian T-bills (with yields of 12.7-13.4%) and Hold recommendation for Ethiopia 24s (with a yield of 6.4% (YTC) – z-spread 604bps – on a mid-price basis as of cob 21 October on Bloomberg).

You can find a full summary of our Egypt meeting here and our Ethiopia meeting here.

Ghana and Bolivia: Fiscal woes and elections

In terms of GDP outlooks, Ghana and Bolivia are quite different. The IMF has revised Ghana’s growth down to 1% in 2020, compared with the 1.5% expected in April, although the extent of the downward revision is not as bad as seen elsewhere and Ghana is still expected to see positive growth. Bolivia on the other hand is seeing a much deeper recession, with a much sharper downward revision to this year’s growth forecast, c5ppts since April, from -2.9% to -7.9%, due to the drop in demand for natural gas exports, leading to a 10ppts swing from 2019.

But the two countries have something in common – a sharp widening in budget deficits this year, and both in election years. Both governments expect double-digit deficits – 11.4% in Ghana and 12.1% in Bolivia. Both will need to produce more realistic budgets for 2021 and medium-term fiscal plans. Consolidation will need to start next year, and hinge on spending cuts and domestic revenue mobilisation. And we think neither has the luxury of delaying consolidation in the manner of the IMF’s new mantra about not removing stimulus too quickly for those countries with borrowing space – neither appear to have much of that. Fiscal rigidities may make that challenge even harder. That said, both country’s fiscal plans may carry some element of electioneering – and Bolivia’s budget process imposes constraints on the interim government – which may unwind after the elections. We might only know the true fiscal commitment when the new administrations take over (on that front, Bolivia’s election on 18 October looks to have returned the socialist MAS party to power with the election of Luis Arce). Availability of financing, and financing constraints, will also be a factor in determining the pace of consolidation.

We have a Hold on Ghana and Bolivia. GHANA 29s yield 9.0% and BOLIVI 28s yield 8.0% (YTC) on a mid-price basis as of cob 21 October on Bloomberg. Yields on BOLIVI 28s bonds are, however, 100bps wider since the election (the bond is down 4.5pts).

For our recent research on both countries, see Ghana: Slow pace of fiscal consolidation a key risk post-election, dated 6 October, and Bolivia: Uncertainty continues ahead of October election re-run, dated 25 September.

Senegal and Cote d’Ivoire: Safe havens

Senegal and Cote d’Ivoire have, in our view, been relative safe havens during Covid, helped by stronger fundamentals coming into the crisis that enabled them better to absorb the shock and allowed polices to respond appropriately. Chief among these solid fundamentals is robust growth, with real GDP growth averaging 6.4% in Senegal and 7.6% in Cote d’Ivoire in the six years to 2019, and fiscal discipline anchored by the West African Economic and Monetary Union (WAEMU) convergence criteria. Both countries had fiscal deficits last year that were in line with the 3% of GDP regional convergence criteria (Cote d’Ivoire at 2.3%, and Senegal at 3.1%, excluding SENELEC – 3.8% including it).

One difference, however, is Cote d’Ivoire’s lower leverage relative to Senegal. Public debt in Cote d’Ivoire is 42% compared with 65% in Senegal. There is also the small matter of Cote d’Ivoire’s elections at the end of this month.

However, the growth impact of Covid has differed this year between the two countries. Both are expected to see a sharp slowdown and both saw further downward revisions in the latest WEO. Cote d’Ivoire will still see positive growth, however, although growth has been revised down to 1.8% this year from the 2.7% expected in the spring. That’s a fairly modest revision. Senegal on the other hand is expected to fall into recession, with growth set to be -0.7% this year, compared with the 3% growth envisaged in the spring, resulting in a big 6ppts swing compared with 2019. Both are expected to see strong rebounds in 2021 (towards 5-6%).

Cote d’Ivoire may be showing more resilience because of the structure of its economy, being less dependent on a single commodity or sector, whereas Senegal depends more on remittances and the travel and tourism sector.

A widening of budget deficits, beyond the 3% convergence criteria, has been permitted this year. Regional leaders in the WAEMU agreed to use the escape clause after Covid hit, and because the impact of the crisis has been somewhat stronger than expected, agreed in late September to a more gradual fiscal adjustment path, putting back the 3% regional target by one year to 2023 rather than 2022 as initially envisaged.

Still, both Cote d’Ivoire and Senegal have been able to contain the fiscal impact, seeing only a relatively mild fiscal deterioration due to Covid compared with many other smaller EM and frontiers (although the deficit is doubling from c3% to c6% of GDP, it is coming from a lower base). Cote d’Ivoire’s fiscal path sees the deficit falling from 5.9% this year to 3% in 2023, while Senegal’s sees the deficit falling from 6.3% to 3% over three years. Consolidation begins next year, albeit mindful of the need to support the economy and not to kill the recovery. Domestic revenue mobilisation will be a focus in both cases, while policymakers will also need to be cognisant of financing constraints – limits to regional market capacity and external financing.

Senegal remains committed to its IMF policy coordination instrument (PCI), an unfunded arrangement. Staff completed a virtual mission in September ahead of the planned second review mission in November.

In Cote d’Ivoire, staff-level agreement on the combined seventh and eighth reviews of the joint extended credit facility (ECF)/extended fund facility (EFF) was reached earlier this month. It is expected to go to the IMF Board in early December. The question will be what happens at the end of this programme when it expires the same month. A successor arrangement will likely be preferred by markets, although the decision will be for the next administration (after its elections on 31 October).

We have a Buy on Senegal and a Hold on Cote d’Ivoire. SENEGL 48s yield 7.1% (z-spread 595bps) and IVYST 33s yield 6.1% (z-spread 525bps) (YTC) on a mid-price basis as of cob 21 October on Bloomberg. We think a smooth election in Cote d’Ivoire would be a positive catalyst for the bonds, even at current valuations.

For our recent research, see Cote d’Ivoire: Tensions mounting ahead of crucial election, dated 9 October.

Mozambique versus Tajikistan

These two seemingly disparate countries share some common features. Both have mega investment projects (Mozambique’s LNG and Tajikistan’s Rogun hydroelectric dam). Both have been hit by the Covid crisis to varying degrees – although neither appear to have suffered as much as others in macroeconomic terms. Both are interested in an IMF programme. Both countries’ sole eurobond yields over 10%. For both countries, the big challenge is getting from A to B (ie from here to project completion and the arrival of significant government revenues). But we think it is easier to see a path for Mozambique than Tajikistan.

Mozambique requested an IMF programme earlier this year, likely in the form of an ECF in our view, even before the full onset of the Covid crisis. The request still stands, although it needs to be dusted off and re-assessed in light of events, and Mozambique received emergency rapid credit facility (RCF) financing from the IMF in the interim. Mozambique also received debt service relief under DSSI. The authorities will need to evaluate conditions and policies for a new programme, although performance and the policy response to date has been encouraging. But governance will surely be a big focus, as will the authorities’ intentions for a SWF (the central bank recently published a proposed model for saving LNG wealth targeting US$96bn). Growth is seen as a bit weaker than expected this year (the IMF downgraded real GDP growth to -0.5% this year in the October WEO compared with the 2.2% expected in April). However, in the near term, the fiscal financing picture looks manageable despite the wider deficit. The budget deficit is expected to reach 7.1% of GDP this year according to the WEO (compared with near balance last year), although this is covered in large part from substantial deposits from the CGT receipts and other donor flows. Consolidation will be needed over the medium term, however, especially as public debt is expected to remain over 100% of GDP until 2025, although it seems the authorities are not simply content to sit on their hands and wait for LNG to arrive.

On that subject, Total’s financing deal in July was a good sign, although Exxon’s final investment decision (FID) has been pushed back to next year (see here). The fiscal deficit is seen falling to 5.3% in 2021 and further to 1.0% by 2024. However, balance of payments financing needs may be more of an issue now than they were a year ago. The crisis has eroded some of the buffers the country had built up since the hidden debt saga. Donor support is a key factor until LNG revenues arrive and it is hoped an IMF programme will help catalyse more donor funding. We think the stamp of approval of an IMF programme would also be a catalyst for the bonds.

Tajikistan has also expressed interest in an IMF programme (as mentioned in the staff report for the authorities’ request for an RCF in May), likely an ECF, although we don’t think a formal request has been made. The authorities will need to look at what policies will support that; and nothing really changes with the unsurprising re-election of President Rahmon in the country’s presidential election on 11 October (and which, unlike Belarus, hasn’t been contested in the country). A revised budget was passed in June in line with RCF commitments, although other measures and a reform package will be needed. Tajikistan has fared better than some, with the economy benefitting from the government’s Covid-response measures, including supply-side and fiscal measures. It also received debt service relief under DSSI. The IMF has revised up its growth forecast for this year as the slowdown is not as sharp as expected – the Fund now expects 1% growth this year, compared with -2% at the time of the RCF, although this is still well below the 7.5% rate last year. Reserves have been supported by higher gold prices despite some current account deterioration due to lower remittances. The budget deficit is expected to widen to 6.0% of GDP this year, according to the WEO, up from 2.1% last year, before declining to 4.4% next year and down to 2.6% in 2022. Tajikistan is not so highly leveraged, with public debt expected to be 48% of GDP this year (in the WEO), although is assessed to be at high risk of debt distress (albeit sustainable on a forward-looking basis). But, despite all this, we think it is the Rogun dam that – although the biggest “prize” for Tajikistan – could also be its biggest problem. The project is high priority and the government is looking to attract investment. Yet, given its size (project cost estimated at c50% of GDP), we think lack of visibility on the costs and financing, and uncertainty about offtake arrangements and the project timeline, is a macro-critical issue and one that is intrinsic to any debt sustainability assessment (DSA). We think it will be difficult for the Fund to conclude any kind of DSA – and a programme – without the necessary disclosure of information and degree of transparency about the project that has hitherto been lacking.

We have a Buy on Mozambique and a Hold on Tajikistan. MOZAM 31s yield 10.2% and TAJIKI 27s yield 12.0% (YTC) on a mid-price basis as of cob 21 October on Bloomberg.  

Ukraine: Programme wobbles

Ukraine sees a slight upward revision to its growth forecast this year, by 50bps from -7.7% to -7.2%, although this is somewhat academic compared with the more worrying signs over the direction of the reform agenda that is putting IMF programme compliance at risk.

It is now reasonably well known that the IMF’s programme review is delayed. Over a month on, the first review of Ukraine’s stand-by arrangement (SBA), approved in June, is still pending as the government’s reform commitment, and commitment to programme objectives, has come into question following specific actions that we see as moving in the wrong direction. Nor is it clear in our mind how long it is going to take to resolve these issues in order to complete the review. This is particularly problematic in light of the country’s large funding needs. We retain our Hold on the US$ bonds.

For more detail, see our separate note, Ukraine: Another IMF review, another delay – notes from virtual meetings, dated 21 October.

DSSI 2.0

DSSI was a discussion topic in several of the calls we had on LIDCs. To no great surprise, the G20 confirmed during the Annual Meetings a six-month extension to DSSI (1 January-30 June 2021). However, this is shorter than some may have wished, although the G20 did concede to consider a further six-month extension if need be at the time of the 2021 IMF/World Bank Spring Meetings.

In our discussions, we heard repeated concerns that some eligible countries have had difficulties with getting approvals from non-Paris Club G20 creditors and that the process has been too slow. The G20 sought to address some of these concerns in its statement through its emphasis on full and transparent bilateral participation (read: China).

For more detail, see our separate note, G20 agrees to extend DSSI, with bilaterals facing more scrutiny, dated 16 October.