SA Fixed Income Insights – Feb 2019

The most significant macro event since our last publication has been the speed and extent of the Federal Reserve’s (Fed) dovish tilt at its January Federal Open Market Committee (FOMC) meeting. The Fed announced that it will be monitoring incoming data and it is no longer committed to a path of gradual increases in the Fed funds target rate as we have grown accustomed to since late 2016. The effective funds rate back then was near 50 basis points and has subsequently risen to near 250 basis points. Instead they surprised by saying their next move could be either up or down on its policy rate and surprised markets by saying it could be more accommodative in terms of its balance sheet policy by ending the balance sheet run-off (quantitative tightening (QT)) sooner than previously communicated.

Risk assets reacted with optimism and this Fed shift re-ignited emerging markets (EM) assets with the carry trade theme coming firmly back in vogue. The rand celebrated posting the best all-in returns within the EM FX universe in the month of January by notching up over 7% total return (spot and carry returns). The J.P. Morgan Emerging Market Bond Index spread tumbled almost 60 basis points since the turn of the year into early February and South African (SA) yields echoed this sentiment by declining
30 basis points from 9.38 to 9.08 on the SA constant maturity 10-year government bond. The overall decline in yields has resulted in a 2.9% all-in return in January for the All Bond Index (ALBI), which brings the total 3-month return for ALBI since the EM relief rally that commenced in late November to 7.5% in rand terms. South Africa local markets have been the best-performing component of the GBI-EM local markets index so far this year for US dollar based investors (Chart 2). With near double digit returns in US dollar for SA Government Bonds (SAGBs) outperforming most EM peers and idiosyncratic risks rising around Eskom support and upcoming budget review (February 20), NERSA tariff decision (March 14) and Moody’s ratings update (March 29), we have opted to turn neutral on duration.

With the bullish views expressed since late November on SAGBs and the rand, and reiterated at the start of the year, and our yield targets of reaching levels close to 8.5% on the R186 benchmark bond, we have opted to turn more duration neutral pre-budget and other key event dates mentioned above. We believe more caution is warranted on both the rand and SAGBs in the near term due to less favorable risk-reward and also because the financial support and operational turnaround and cost cutting initiatives combined with NERSA tariff application needs to be assessed in order to make a better judgement on whether Moody’s is likely to revise its credit ratings outlook from current stable to negative. While we do not have a high conviction on the Moody’s ratings pronouncement and wait for further clarity on what policy initiatives will be taken on Eskom subsequent to the State of the Nation Address (SoNA) announcement that the company will be vertically unbundled into three separate units, we do currently see higher risk of Moody’s lowering its ratings outlook on SA sovereign. We expand on this in the section (The risk of a negative credit rating decision from Moody’s has increased).

The market read into the Fed’s January monetary policy statement and press conference  that the Fed is likely done with the hiking cycle (or at very least in an extended pause before the hiking cycle resumes). United States (US) short rate pricing such as Fed funds futures and Eurodollar markets are no longer pricing any rate rises for 2019 with pricing flat in terms of the remaining 2019 meetings. The market is starting to price increased odds of the Fed cutting rates as early the first quarter of 2020. Moreover, the January tone of the Fed and comments by Chairman Jerome Powell suggests that a plan will be forthcoming on how the halting in the balance sheet runoff at upcoming meeting will be handled. We mention later in this article comments by Philadelphia Fed Patrick President Harker that made insightful comments on the Fed balance sheet. This provides a much-needed glimpse into what could be communicated by the Fed in the meetings ahead. Most likely the plan will involve the current QT of $50 billion/month could be tapered in coming months and ultimately ending potentially as early as the fourth quarter of 2019 or quarter one of 2020, leaving the fed balance sheet at a larger level than previously anticipated and higher level of excess reserves in the banking system.

The January Fed meeting outcome is consistent with the views expressed in the Fixed Income monthly insights piece when we hinted that we expected a dovish tilt by the Fed may be lying ahead in the first half of the year. A prescient view on a policy shift and its implications for EM and by extension SA rand and bonds allowed our funds to benefit from this key macro development. We have opted to start taking profit on this to monetise our views after the strong performance by bonds in early 2019. Investor surveys suggests that EM long positioning has risen substantially and has led us to be more conservative in our portfolios as the marginal buyer of EM assets could be somewhat curtailed.

It has been quite remarkable that the Fed has done a 180 degree turn from their prior meeting held five weeks prior to its most recent FOMC meeting. This decisive shift by the Fed from its former hawkish policy of gradual increases in the Fed funds rate with dot plots rising above 3% in the long run, has provided a needed lift to risk assets with EM benefitting and EMFX benefitting from a return in carry. The next dot plot and economic projections are only due at the FOMC meeting taking place late March, which will provide more clarity on the Fed’s thinking of its rate path and economic outlook. The January statement referenced the words ‘crosscurrents’ in the economic outlook and the Fed’s pivot was likely induced by the market tantrum and weakness expressed in the latest US household surveys conducted.

The January FOMC statement is significant as it likely marks the start of the end of the cycle that commenced two Fed chairs ago when the Ben Bernanke Fed signaled end of easy money that sparked the Taper Tantrum in May 2013 when Fed tightening was signaled. With the Fed likely to announce slowing and ending of QT, this should provide a positive boost to risk sentiment and liquidity in the months ahead. But investors need to be wary, while this announcement is expected to occur as 2019 unfolds, other issues such as US-Sino trade talks, US political risks surrounding the debt ceiling and potential return to partial government shutdown and global growth slowdown fears together with Brexit concerns are some of the factors that are more threatening to markets in the shorter term. As things stand, markets have enjoyed the People’s Bank of China liquidity injection, which together, this Fed policy shift in January versus December has resulted in Global Money Supply jumping to levels last seen in March last year which has arguably been the main underpin (as opposed to macro variables or corporate earnings) for the huge rally in markets from Christmas eve doldrums. (Chart 1)

Chart 1 – Global Money Supply surges over $2.5 trillion since late November with the bulk occurring post-Christmas 2018

Screen Shot 2019-02-18 at 15.48.43
Source: Bloomberg

For now, we wish to highlight that the Fed is still pursuing a balance sheet runoff. If the Fed continues QT for some time to come still, not only will $200 billion in liquidity vaporise every four months but de facto tightening will amount to one hike a quarter, financial conditions will tighten as if Fed had maintained a “gradual” rate tightening pace. It is therefore critical to the monetary policy outlook and gauging the market response as the Fed decides how to tweak its current balance sheet policy and wording.

The Federal Reserve Board’s dovish turn has grown more explicit with each passing week thus far in 2019, punctuated by last week’s FOMC board meeting; risk has been recovering on the back of this. The broader rates backdrop has therefore become more benign, favouring currencies more heavily reliant on debt capital flows. Higher yielders have been the biggest winners, and we expect this dynamic to persist in the medium term but can be punctuated by domestic idiosyncratic factors that interrupt rallies.

Chart 2 – GBI-EM Country Returns year-to-date 2019 (South Africa)

Screen Shot 2019-02-18 at 15.22.43
Source: Morgan Stanley

The big uncertainty is the growth backdrop. The deceleration in economic activity data out of Asia, and China especially, has become a primary concern for investors. There is ongoing doubt about the scope and efficacy of China policy makers’ stimulus efforts and how long it will take to bear fruit (Chart 3). We remain more constructive than the consensus on this front. Combined with progress on trade talks and the positive knock-on effects from a more dovish Fed, we believe that the EM rally has legs into the second quarter of this year and beyond. The reduction in political risk in in the second half - 2019 after a busy first half schedule of elections in Asia and elsewhere could provide an additional boost to EM assets.

Chart 3 – China Credit Impulse and China Purchasing Managers’ Index relationship suggests it could take more than two quarters for full effect of recent policy stimulus to reflect in data

Screen Shot 2019-02-18 at 15.38.37

Source: Bloomberg and Saxo Bank

The rand seems to have encountered a challenge to its bullish momentum in the past week. The dollar/rand traded above 14.50 as recently as December. Since then, the fears that the US economy was slowing has caused a surge of cash to flood towards EM in a renewed “search for yield”. Various indicators, however, suggest that this EM rally may be faltering. Firstly, the latest US data – think the most recent US payrolls and ISM manufacturing bounce back - points to an economy that is still fairly robust and so it may just be a matter of time before the Fed may be altering language that further hikes are not off the table completely and then investors no longer have to look towards EM for returns. Secondly, various measures of investor sentiment – from positioning to pricing measures such as volatility – are already at stretched levels that have usually triggered market turnarounds. Thirdly, the rally in EM yields seems to have run its course and even currencies have started battling since early February. Fourth, dollar/rand faces very strong support to break below the 13.25/30 level.

There are also numerous risks ahead – only one of which would have to prove problematic for the EM/rand rally to falter. These event risks range from global issues, such as US politics, Brexit, trade wars, to local factors, such as the upcoming Moody’s review, the budget, Eskom and so on.

The SoNA was well received by markets, though there is a material risk that the solutions presented for Eskom’s crisis might appear positive but may have ambiguous consequences for SA’s sovereign credit ratings. Moody’s indicated after Eskom’s request for substantial fiscal support, including a R100 billion debt transfer, that the rating impact that this will have is not clear-cut as it depends on the other components of its turnaround strategy. Our analysis of the Moody’s sovereign credit rating methodology and its SA assessment still suggests that such a debt transfer would significantly increase the risk of negative sovereign credit rating action by Moody’s, though we concur that it would not be a done deal (see section discussing Moody’s most recent comments). The rand doesn’t seem to adequately reflect this uncertainty and the other lingering negative risks.

Commodities are mixed, but iron ore chalked its best level since 2014 as the crisis at Vale intensifies. In SA, with three months to the elections, President Cyril Ramaphosa made an investor friendly SoNA. It was revealed that Eskom will be split in three (government financial support to be disclosed in the 20 February’s budget); economic zones will be set up for the clothing industry; a crime busting investigative unit will be formed and report directly to the National Director of Public Prosecutions; electronic visas will be introduced; and government will tackle hindrances to doing business.

Next, we await Minister Tito Mboweni’s delivery of the 2019 budget speech on 20 February. Market hopes are high that he will show a fiscal trajectory with meaningful improvement compared to the October 2018 Mid Term Budget Policy Statement. Unfortunately, we see the fiscal deficit remaining stubbornly wide, given mediocre economic growth and only limited room for expenditure cuts. This keeps the risk of a Moody's outlook downgrade alive, particularly if the national government takes on the responsibility of propping up Eskom's balance sheet.

The risk of a negative credit rating decision from Moody’s is high

We judge the risk of a negative credit ratings announcement (an outlook revision to negative from stable) as early as the Moody’s scheduled announcement on 29 March as high. However, Moody’s could decide to be pragmatic and wait until after elections to make a decision and/or officials could surprise us all and make bold policy announcement (e.g. on Eskom) ahead of the election in May.

Moody’s ratings methodology maps SA to Baa2, bordering Baa3 on both sides (Chart 4). The agency’s official rating has previously been the average of adjacent ratings on their grid, hence the current Baa3 rating. Material slippage in any of the four key factors: economic strength, institutional strength, fiscal strength or susceptibility to event risk, would see the rating map to a Baa3 bordering Ba1, likely resulting in a downgrade to Ba1.

There are four ratings factors and we list them below together with the risks around each:

  1. Economic strength: Lower realised and expected gross domestic product (GDP) growth will put negative pressure on this ratings factor. However, announcements of credible structural reform efforts could lead to positive ratings adjustment scores, buying the sovereign time. There is high risk of slippage in this respect, and we would be looking for concrete announcements soon to avoid a downgrade.
  2. Institutional strength: The developments to watch here are accountability around state capture and corruption, followed by how policy formulation around the South African Reserve Bank’s mandate, land appropriation without compensation and prescribed assets shapes up. These proposals were included in both the ANC’s policy resolutions and the election manifesto. Let’s hope the country does not shoot itself in the foot as far as these issues are concerned. Ability to meet budget targets might also be a drag on this factor. There is moderate risk of near term slippage here, but it is a risk in the long term.
  3. Fiscal strength: The key risk here is Eskom. Moody’s has said that how government resolves the Eskom issue will be as important as whatever transfers are made to the utility. The President has promised that Eskom will release a turnaround strategy in the “coming weeks”. Clarity on Eskom cannot come fast enough, but we worry that government will not have appetite for unpopular announcements ahead of the election on 8 May).
  4. Susceptibility to event risk: This is the one factor that might be stable at a high level due, in part, to the strength of the banking system, depth of domestic financial markets et cetera.

Chart 4. Moody’s ratings factors and rating history



Screen Shot 2019-02-18 at 15.24.17
Source: ABSA

Moody’s Ratings history on SA Sovereign Foreign Currency Long-term Debt

Screen Shot 2019-02-18 at 15.38.57

Source: Bloomberg

Key Points on Moody's comments after SONA

• Highlights the lack of reference to “concrete measure to overcome structural challenges in the face of entrenched vested interests”.

• Eskom is a problem, but as previously noted, financial support is (1) credit-neutral for the sovereign if accompanied by measures to stabilise Eskom’s financial health; (2) credit-negative for sovereign if support is provided upfront, but cost savings and/or tariff increases come much later (since those require unpopular decisions). “The overall credit implication of any financial support for the sovereign will hinge on whether or not any support is part of a set of broader measures”.

• Moody’s does not expect “any marked acceleration in growth in the next few years”, although the decline in growth potential might be arrested by the announced government policies. Their views for growth are around the 1.4% mark for 2019 and 1.5% in 2020. The risk is that protracted load shedding causes this number to fall significantly and load shedding by itself if unaddressed could imperil SA’s investment grade rating.

Some further points from Moody’s made at its most recent client call regarding their thinking on Eskom and SA sovereign risk include:

1. The SoNA offers a realistic diagnostic of the economy and a commitment to reform. That gives them comfort. Having said that they feel that turning the economy around will take time. There was also limited guidance on concrete measures to create jobs, raise competition, reduce inequality etc. Commitment to addressing these issues is positive, but for now leaves them quite cautious.

2. Financial troubles at State-Owned Enterprise’s make it difficult for government to balance job creation objectives with fiscal discipline.

3. Financial support to Eskom is seen in two ways: i) size of the support and ii) what it does to the health of the company, 2% of GDP debt transfer is significantly above what normally transpires when Eskom gets support.

4. Regardless of the form of support, if such measures come with credible and likely steps to address the Eskom cost base in the near term, that would be credit neutral. If the support package is accompanied only by steps to reduce the cost base at some point in the future, that would be credit negative, as it carries higher execution risk.

5. An equity injection would probably leave contingent liability exposure unchanged, so while Eskom may be in a better situation, the sovereign still carries a very large risk on its balance sheet.

6. The Sovereign analyst didn’t want to be drawn on the required time lines for the Eskom turnaround plan to bear fruit, or what that plan would look like. Only that i) contingent liability risk is front and centre for them, and ii) there is risk with support coming first while management of contingent liabilities comes later. They will look at Eskom turnaround plan and make their own mind up on whether it is likely.

7. Moody’s can move on any date, but action, if any will probably only happen on the scheduled dates in March and November. The most important metric they are watching is the level of debt and whether or not it stabilises over the Medium-Term Expenditure Framework (MTEF).

The size and treatment of Eskom has very different implications for the fiscal metrics depending on which way it is delivered. An equity injection poses a risk to the expenditure ceiling, while a debt transfer widens the budget deficit over the MTEF and raises the debt ratio. For now, our base case remains that Moody's will be on hold on March 29, but the risk of an outlook downgrade has risen materially. A cash injection would be perceived fiscal neutral, if this cash comes from government non-core asset sales (e.g. equity holdings) and would be the best option in our view combined with cost cutting, an operational turnaround plan and reasonable tariff increases in the 10-15% per annum range. The size and treatment of Eskom has very different implications for the fiscal metrics depending on which way it is delivered. An equity injection poses a risk to the expenditure ceiling, while a debt transfer widens the budget deficit over the MTEF and raises the debt ratio. For now, our base case remains that Moody's will be on hold on March 29, but the risk of an outlook downgrade has risen materially.

We see near term risks on the horizon for rand that keeps us defensively positioned

Last month we expressed a weak US dollar thesis and expressed an opinion that rand could be a beneficiary of the Fed policy stance change, stronger commodity prices, EM flows and from US Treasury drawing down its cash reserves at the Fed hat creates more supply of US dollar in the system. The dollar/rand was able to move into the 13.00-13.50 range as we anticipated trading as strong as 13.2380 on 31 January.

Subsequently our view has changed in the shorter term due to increased risks around Eskom headlines, the budget and investment flows related to corporate activity. Upcoming corporate unbundling activity can potentially lead to equity outflows by foreign investors and bring a negative flow impact to the currency.

Naspers is due to spin off MultiChoice Group (MCG) to its shareholders on 27 February which will occur on the Johannesburg Stock Exchange. The ownership structure suggests that foreign funds could own 60% of MCG, split between MSCI EM-indexed investors and global tech-focused funds. The latter group could be potential sellers of MCG following the unbundling and separate listing, since their mandate would not involve ownership of this type of stock.

Multichoice is likely valued in the R65-79 billion/US$4.8-5.8 billion range which is placing MCGs equity at R150-180 per share. This estimate is based on target the financial year2020 earning per share estimate and using an EV/EBITDA multiple of 7-8x, in line with global peers. If foreign selling results in 20%-25% of liquidation it could result in $1-2 billion off equity outflows in the month of March.

Besides the flow factors there could also be an FX impact driving dollar/rand higher from foreigners hedging local SAGB holdings, if they see Moody’s ratings action as a threat to their South Africa exposure in their portfolios. We turned more bearish on rand when dollar/rand was in the 13.25-13.50 range and would feel more comfortable to turn less defensive on rand at levels of 14.40 or above. With dollar/rand implied vols very attractively low at the moment we see value in FX vo. The rand tends to appreciate in the weeks leading up to elections hence we will use weakness in coming weeks to likely increase rand exposure again to benefit from tactical rand longs (dollar/rand shorts) around and leading into the elections set for 8 May. This is also predicated on the assumption that Eskom can be stabilised and Moody’s ratings action averted ahead of the elections.

It is clear just how far the pendulum has swung towards optimism when looking at ranges and changes to the dollar/rand FX vol surface in recent years. About one year ago we were experiencing the height of “Ramaphoria” along with a relatively favorable global backdrop. Therefore, the curve from one year ago could represent a benchmark for how low vol could go. Today’s volatility surface looks almost identical to the low and flat volatility surface that marked the peak of optimism when rand strength saw dollar /rand test levels near 11.50 in March 2018. The 3-month tenor for then and now are essentially the same, the overall average is nearby, but we are steeper now, mainly I guess because it is an election year and that risk event falls just past the
3-month peg. There is the endless list of geopolitical risks we currently face, none of which necessarily wait for more than a month or two to cause problems. Interestingly, 10 delta US dollar calls (dollar/rand 15.16 strike) are trading at similar vols to an ATM curve of 3-years ago when we were in the stressful times of President Jacob Zuma firing and hiring finance ministers. Their prices have plenty of potential to explode higher if just some stress enters the market. A three month 10delta US dollar call (USDZAR 15.16 strike) today trades off a similar vol to what a three month  ATM traded 3-years ago. So, if there is a shock, the ATM increase alone may be enough to justify the vol of those low delta Calls, let alone increases in skew if those options remain OTM. The risk/reward on long EM assets has deteriorated. There are mixed signals coming from some of the technical/quant indicators we look at, while sentiment has turned bullish. This is not the time to be adding additional long exposure to EM in our view.

Further thought on the Fed Balance sheet policy. As expected, the Fed updated its policy normalisation principles and plans, including balance sheet policies. It committed to the floor-based system with abundant excess reserves for an “indefinite period” and reiterated its view that interest rate policy will be used to fine tune the policy stance and balance sheet policy will only be adjusted to the extent that interest rate policy is insufficient to properly calibrate the policy stance. Chairman Jerome Powell said that the committee would be taking decisions at upcoming meetings about what plan the committee has about winding down the balance sheet run-off. He seemed to imply that there would be a gradual phasing down of cap sizes but was not overly explicit about the committee’s plans beyond signaling that further announcements are forthcoming.

Philadelphia Fed President Harker said recently in an interview that the Federal Reserve is close to announcing an end date for its balance sheet reduction program and will decide on the composition of its securities portfolio afterwards. He said that minimum  Fed Reserves to cap volatility maybe $1-1.3 trillion plus a buffer to cover fluctuations in non-reserve liabilities that can exceed $100 billion monthly. He indicated that the Fed will also think about the composition of its balance sheet once the decision has been made on level of reserves.

With reserves currently at $1.6 trillion and shrinking by up to $50 billion a month, the wind-down may need to end as soon as this year, leaving the Fed with a much larger balance sheet than prior to the financial crisis. The normalization principles suggest that the Fed will eventually shift to a portfolio primarily of Treasuries, but officials continue to debate how quickly they should shed $1.6 trillion worth of mortgage-backed securities. 

Another debate centers on the duration of the Fed's Treasury holdings. Operation Twist in 2011 saw the Fed selling short-term securities and buying longer-term bonds to lower longer-term interest rates. Now it holds $2.2 trillion in notes and bonds but no T-bills.
 Analysts have suggested that a shorter duration portfolio would allow the Fed to extend duration as a way of adding monetary easing in a recession. 

The balance sheet remains a passive tool, even as the FOMC last month signaled openness to adjust the pace of normalisation if necessary. Harker said the revised guidance marked a minor change from the FOMC's stance that the fed funds rate is its primary means of adjusting policy.

When asked at the press conference how the Fed would determine the terminal level of reserves. Chair Powell responded: “There is no playbook. No one really knows. The only way you can figure out is by surveying people and market intelligence, and then ultimately by approaching that point quite carefully”.

Powell mentioned “surveys and market intelligence” earlier in the press conference and with that in mind the New York Fed’s Surveys of Primary Dealers (23 counterparties of the New York Fed) and Market participants (32 institutional investors) will take on greater importance as the Fed feels its way through balance sheet normalization. The fact that Powell specifically referenced these surveys as a metric for guiding future balance sheet policy is key and suggests future surveys will be worth watching closely.

In closing, we maintain our cautious stance on local SAGBs exposure but will be willing to move back to overweight duration at levels that meet or exceed 9.15% on the R186s if yields were to start blowing out on the current idiosyncratic risks. We expect near term support near 8.98% on R186 but would rather sit on the sidelines and use overshoot opportunities in yield backup episodes to increase exposure to long-end of the curve. The market is starting to price risks better surrounding Eskom in recent days with yield having backed up above 8.90% on R186. Should the Eskom turnaround story be credible and financial assistance extended in such a way that Moody’s deems not credit negative, we expect some retracement from recent highs in bond yields. In the medium term, we expect our local yields to benefit from low/falling core bond yields in G7 economies from disinflation from China being exported and sluggish or deteriorating global growth remaining a key market theme. We are therefore looking to fade extreme bond weakness if selling prevails around concerns over Moody’s ratings action.

We will also allow our decisions to be guided by increase in governments borrowing requirement in the budget which we think will be somewhat contained and would be surprised to see material increases in borrowing requirements at the 20 February Budget Review. Uncertainty around the National Health Insurance (NHI) funding that may be disclosed could be a source of negative surprise for bond investors that needs to be considered given that the President has stated that the revised NHI bill will be submitted to parliament later this year.