Friday, 20 August 2021

Ethiopia's likely coming bond default

If you follow the news then you know what has been happening there, if not this is a cliffs note summary:

The EPRDF coalition political party ruled Ethiopia with an iron rod from 1991 to 2018 with the Tigrayan Peoples' Liberation Front, TPLF, faction being the most powerful. Federal power based along Ethnic/ tribe-based Federal states kept the ethnic/ tribal differences and conflicts subdued. PM Meles Zenawi, and after he died, his successor PM Hailemariam Desalegn, ran the Federal state as a technocratic economic development project and the economy grew about 10% a year for many years.

But the largest ethnic group, the Oromo and the Oromia Liberation Army, felt subdued by the prior Amhara kings and then the EPRDF, which was dominated by the Tigrayans. (Amhara is the second biggest state and created modern day Ethiopia via the prior Abyssinian Empire via the invasion of the surrounding tribes). 

Protests and violence came to a head in 2018 and via a parliamentary reshuffle PM Abiy Ahmed, whose parents are both Oromo and Amhara came to power. Abiy was initially greeted as the new face of Africa - a young, reform minded, peace maker. In 2019 he was awarded the Nobel Peace Prize for securing peace with Eritrea. 

However, the TPLF did not support him and the EPRDF party split into the TPLF and Abiy's new Prosperity Party. 

Anti-corruption drives and other measures were viewed as an Abiy-led purge against the TPLF and things came to a head in late 2020. 

Abiy sent 300 special forces to Mek'ele, the Tigrayan capital, who attempted a night time raid on TPLF leaders on the 3rd-4th November 2020. This failed and the TPLF and their paramilitary wing, the TDF, launched a counter attack on the Ethiopian National Defence Forces Northern Command bases on the 4th November.

This led Abiy to declare war on them and initiate a prepared invasion with the ENDF, and he was supported by the Eritrean Defence Forces, Amhara Special Police and crucially drone support from the UAE, who had a drone base in Eritrea, plus a few hundred Somali Army trainees. 

As has been documented the UAE drone support destroyed the TDF's heavy weapons and they quickly abandoned a conventional warfare strategy and melted in the mountains. Bringing retired military commanders out of retirement, most notably General Tsadkan Gebretensae, they then trained over 100k volunteers, while carrying out hit and run attacks on the invading forces. 

The invading forces adopted pretty much a scorched earth policy and all sides have been accused of numerous war crimes. 

The first turn in favour of the TDF was in February the UAE was forced by the US to disengage, which left the ENDF without much aircover. 

Then at some point in May the Eritrean troops stopped engaging fully in the new attacks. This left the ENDF exposed to concentrated TDF attacks. The ENDF mounted two failed efforts to take the highlands and lost thousands of troops in the process. The TDF had also been building troop numbers and launched Operation Alua, which led to the total collapse of the ENDF in Central and Southern Tigray and the recapture of Mek'ele, the state capital. Eritrean troops withdrew to the northern border areas.   

The TDF then invaded part of the barren Afar region, to the east of Tigray, threatening to capture the Main Supply Route from Djibouti to Addis, but this turned out later to be diversionary and instead the main forces invaded the Amhara region heading towards Debark in the north west and Weldiya in the east. 

After capturing Weldiya the TDF have headed south towards Dessie and west towards Debre Tabor, which has been captured this week. Debark, which sits by a mountainous ridge line, has been so far defended by the Amhara forces.

This month they have also agreed to cooperate with the Oromo Liberation Front's Oromo Liberation Army to bring down Abiy. The TPLF and OLF were arch enemies in the past, so this is a remarkable redrawing of alliances. 

Over the last few months the OLA has grown from a small force into one fielding divisional sized forces and they have been capturing land north west and to the south of Addis.

Wikipedia and Ethiopia Map on Twitter are keeping up to date maps of territory held and lost. 

 


This week the OLA claim they have control of the road down to the Kenyan border and have attacked Suluta, just 12kms outside of Addis to the north.

Simultaneously the TDF are now marching towards Bahir Dar, the Amhara regional capital. 

 

Abiy's regime has gotten to such a desperate state that they seem to have accepted sanctioned military drones from the Iranians and have been asking the Turks for additional military support, but it is likely too little, too late. 

 

So in summary, I think we can see that the Abiy regime is a few key battles away from collapsing and the siege of Addis is now beginning.

 

Ethiopia's only bond

Which brings us to the bond. 

A 10-year $1bn bond was issued in 2014. 


Ethiopia to get to 10% GDP growth ran up 60% debt to GDP, ran CA deficits/ borrowed from abroad and has taken part in the DSSI programme. 

Annual GDP growth:


 
 
Debt/ GDP:

Current Account/ GDP:


Multi-billion projects like the $5bn GERD dam cost a lot of money upfront for years before generating anything yet are key to economic development, for example. 
 
In January 2020 they entered an IMF Extended Credit Facility programme. The IMF expected them to slow growth, reduce current account and government deficits, lower debt/ GDP, engage in economic reforms and build reserves. 

They were then hit with the Covid pandemic and took part in the DSSI initiative to reprofile $1.2bn of bilateral debt servicing cash flows.

They have since announced an intention to reprofile debts on a 'voluntary basis' and have so far indicated that this would not include the Eurobond. 
 
Excluding the Eurobond goes against the G20's 'Common Framework', which calls for all debts to be treated equally.

The biggest holder of the Eurobond, according to Bloomberg's analysis of 13F filings is Franklin Templeton fund manager, Dr. Michael Hasenstab, who shall we say is not unfamiliar with sovereign stress and distress.

 
 
He held a baseline position of $70-90m at least since 2015. It is hard to criticise that, he was supporting a growing, low-income country, with what was initially taken as a new reform minded Prime Minister, Nobel Peace price winner, Abiy Ahmed. 

However in Q1 when the bond prices sold off into the low 90s, he bought more and took his holding up to $130m of par value in total. The main portfolio holding the bond seems to be the Templeton Emerging Markets Bond fund.


According to the fund holdings he took it to a 2% position there.

I am not sure what the Franklin Templeton ESG policy is, but he did this while Ethiopia was financially stressed, asking for debt relief, in an IMF programme and additionally when there were numerous reports of widespread war crimes being committed by Abiy's combined forces in Tigray. 
 
The bonds are currently indicated in the high-80's, but hardly trade. 
 
 
Default and Restructuring
The IMF would typically want a low income country to peak at 60% debt to GDP, net of an IMF programme, yet Ethiopia went into the last 18 months at that level of debt/ GDP. 
 
In my opinion, the country will end up with a restructuring and a meaningful NPV debt haircut.

In my opinion the Eurobond will be included in the debt restructuring. If the new government is quickly recognised by the US and IMF then perhaps a 65c recovery at an 11% exit yield may be achieved by the end of next year. However the bonds should trade significantly below that before and after the restructuring as the negative news hits, perhaps they trade down to 30-40c. 

If the US/ IMF don't recognise the new government then it becomes a hard default against a possibly sanctioned regime, if that happens, then sub-10c will eventually be realised. 

I do think the former outcome is much more likely but the bonds could easily trade under 30c amidst the initial uncertainty and the time frame for normalisation can draw out easily. Post restructuring bonds are often trading at 15-20% yields for at least 6-12 months in many recent instances before cross over investors start to buy back in and yields normalise.

So on a $130m of par, a $100m mark to market loss is not inconceivable with maybe $40-50m in permanent losses, post restructuring, over the medium term.
 
Complicating factors could be the expected secession of Tigray, which would cost Ethiopia 7-8% of GDP. Tigray could also claim war reparations at least against the Amhara state, which represents about 25% of GDP. 

There is also the likelihood that the TDF will go after Eritrea's totalitarian dictator Isias Afwerki and invade Eritrea. Whether that results in a new democratic, independent Eritrea, a merged Tigray/ Eritrea which delivers a defacto secession quickly, or Eritrea is again assumed into Ethiopia, that remains to be seen. I think it is inconceivable that the TDF will make the mistake of leaving Afwerki in power again as it did in 2000 when the Army was ordered to stand down by PM Zenawi.


Friday, 9 July 2021

How far behind the curve is the Fed?

Over the last few months Treasury yields have dropped and retraced a large amount of the Q1 sell off.


Commodity prices have peaked, in my view, stimulus has or is peaking, and the usual suspect deflationists are calling for lower bond yields. 

Set against this is the ongoing economic recovery and, in my experience, without a massive unexpected shock and with the ongoing fiscal and monetary stimulus, its hard to see why the US economy would relapse into a recession and deflation. 

The fact that the US Current Account has widened suggests that there is ongoing excess demand in the economy.

Instead, the drop in US 10yr yields since March is probably more down to the combination of a normal consolidation phase following a large move and then some short term supply/ demand technicals, concentrated positioning, stop losses and regulated buyers vs the fact that the 10yr-plus part of the Treasury market is a relatively small amount of the outstanding, with 10-30 year bonds being about 20% of new Treasury issuance.

Instead we are likely to see a normal reflationary cycle of unemployment falling, capex rising/ service industry job creation, wages rising and ultimately a hiking cycle followed by a recession. 

So where are we in this cycle? 

The Fed today published its monetary policy report. Their ongoing conviction is that yes the economy needs more support in the form of ZIRP and continuing QE and that they are not ready to commit to tapering, but might be later in the year. In fact, over half of the FOMC participants expect Fed funds to have zero to two hikes by the end of 2023 with 5 of those expecting zero hikes! 



Unfortunately for them, the strong recovery in the US economy has now led the number of job openings to be larger than the number of U3 unemployed.

The last and only time this has happened before was in early 2018 when Trump was stimulating the late cycle economy and the Fed was half way through their hiking cycle. US 10yr Treasuries were yielding around 2.4-2.5% and fed funds were at 1.4%. 


So even by the Fed's own track record in the most recent cycle, they are far behind the curve, their projections are absurd and the excess stimulus and Fed balance sheet growth is causing many asset bubbles to extend. 

The excess balance sheet growth in the last cycle never really led to a wage-price inflation spiral. So the Fed never got trapped with a large balance sheet leading to excess reserves in the system that then lead to credit growth and rising monetary velocity. 

When asked about when the Fed would unwind its balance sheet, Bill Gross famously quipped 'never'. And for the most part he was right. 


In this cycle however the starting point is with over 3% wage growth whereas in 2009 it was well under 2%. In the last cycle job switcher wage growth almost hit 5% before the recession. 


In this cycle, given we already have more job ads than employable unemployed, most probably wage growth will exceed 5%.

If that combines with credit growth and a weak Dollar then the Fed will be forced to act much earlier than expected. 

Dec 2022 and Dec 2023 Eurodollar futures have diverged this year but Dec 2023 is only pricing in 3-4 hikes and Dec 22 only 2.

 


As mentioend above the last time the employment market was this strong the Fed was already at 1.40% in early 2018 and by year end had hiked to 2.4%. That sent the 10yr from 2.4% in early 2018 to 3.2% by year end, the cycle high, as in 2019 the economy slowed and manufacturing went into recession.

Fed funds rate:



So if wage inflation starts to rise and point towards higher inflation going into 2022 the Fed will be under immense pressure to speed up policy normalization and hike rates.







Wednesday, 24 March 2021

Inflation? What inflation?

I had a break from posting, partly due to other commitments and partly as I had similar views as what the market was doing, namely looking for a liquidity fueled recovery. 

I did expect a correction in the summer or Autumn, particularly into the Corona re-outbreaks and US election, but in the end the power of stimulus was too much and it was more of a temporary sideways move. 

So in a world of ZIRP, QE, unprecedented fiscal deficits, modest curve steepness, sky high asset values, what can derail this?

We hear a lot about inflation and the impending USD collapse, but the USD has risen this year against the majors on the yield differentials. 

 

Modern inflation

Under the gold standard, with mostly industrial economies, inflation was certainly a monetary phenomenon and the gold standard prevented it. 

As we have moved towards a services economy with open goods trade and a fiat monetary system where credit, the financial system and the money supply can grow faster than the real economy for long periods, there has become really two forms of inflation; firstly money supply/ credit/ financial economy growth and secondly skilled wages.

QE and base money creation cause, by the portfolio channel effect, asset price inflation as investors seek higher returns than available in fixed income. To the extent that, with fiscal stimulus, delivers a recovery then broad credit/ M4 growth also happens. So you have credit supply and asset inflation underwritten by QE/ ZIRP/ fiscal stimulus. I think ZIRP/ QE is most relevant for financial assets while fiscal stimulus is most important for the real economy cycle. 

As Steve Keen has shown, in a financialised economy, private sector credit going from growth to contraction causes a final demand shock and that can lead to a deeper real economy recession and a credit crisis in the financial economy. So governments have stepped into the Covid crisis with unprecedented fiscal stimulus to support final demand. 

This demand support combined with supply chain issues had pushed up commodity prices. However rising or falling commodity prices, while they impact CPI, are really just a relative pricing power shift between goods producers.

So when people predict inflation from QE/ fiscal deficits, is it showing up in credit or the money supply?

The Fed's H8 survey suggests not, while Treasury supply is growing, large parts of private sector credit is contracting and overall bank credit is growing ~6%. So while the inflation has shown up in asset prices, there is no sign of it creating inflation in the real economy. 


 

The Fed's recent projections forecast a 6.5% GDP growth this year on a $21Tn Q4-20 economy, while bank credit is growing about 6% on a $15.6Tn banking balance sheet. So bank credit is currently growing at a disinflationary pace vs the economy. Mainly due to private sector deleveraging. I dont think non-bank credit will be doing much differently given the broad deleveraging by private borrowers. 

Q2 inflation outlook

The Fed is expecting YoY inflation numbers to peak in Q2, which is when the difference in commodity prices will be highest and there will likely still be some supply chain disruptions for goods. I agree and also expect inflation to drop back in the second half. 

The NY Fed UIG inflation prices only metric is at 2.3% and looks to be rising, while CPI is only 1.7%. If you look, CPI ranges and mean reverts around the red coloured UIG prices measure, so should overtake and overshoot to the upside, maybe 2.5-2.7% in Q2 or early summer. The Atlanta Fed's 1yr forward inflation expectation is rising strongly to 2.4%.  





Treasury yields should peak a month or two before all of this, so likely April or May in an upside squeeze. 

The recent topping in oil and renewed real estate foreclosures and rising mortgage yields should also soften owner equivalent rent and normalising supply chains should take the edge off goods price inflation. Service inflation has been sticky at a little under 2%. 

So I think 1.95-2.05% on the 10yr TSY is a reasonable target. It would give almost 200 bps steepness for the 10yr-2yr TSY spread, which is more or less as high as curve steepness gets in recent initial recovery phases before the spread moves into a range. 

 



After that I would expect some of these inflationary pressures to drop and TSY yields should also drop back to maybe 1.40-1.50% on the 10yr and the weak USD/ commodity reflation/ EM reflation trade should be back on. 

Service wages

However, if monetary inflation is not an issue, perhaps later this year going into next year skilled service wages will be. In most recessions wage inflation has reduced by about 2%:


But in this recession it has hardly moved and most of the job losses have been lower skilled jobs. 


Skilled worker employment is only down 2-3% YoY now while lower skilled it is more like 4-5%. The JOLTs job survey is also showing strong job growth, so over this year a lot of the employment slack should go and we get a tight market for skilled jobs. 

In fact the Atlanta Fed's job switcher wage inflation metric which peaked at 4.3% has hardly moved down this time. The last two recessions it dropped 3%.


To going into next year if job switcher wages are growing 4.5% or maybe more, the pace of the recovery is slowing, service inflation is picking up, maybe 10yr yields go for 2.5-3% with the market looking for a hiking cycle starting in mid-22. 

If this is coinciding with fiscal deficit reduction, while equities are likely to rally as the USD weakens in H2, maybe in 2022 if margins come under pressure, the outlook will be for lower equity valuations.

In the end it will be skilled wage inflation that will force the Fed to hike and tighten financial conditions and this wage pressure leading to hike pressure will only end with a genuine labour market recession.  

 











 

 

 





Tuesday, 7 July 2020

A little market update

The Nas pulled back as the leading market on a technical signal. Most of the day was showing a doji reversal signal though but ended the session weak.


S&P looks like it is being banged between the daily chart levels on level 3 until it breaks out in August/ September either way.



But watch the 10yr for confirmation, I think a break below 60bps is key.

Yen from before March (when the US curve inverted QE was repatriated):



Rebased from after:







Friday, 19 June 2020

Where does the bear market rally stand?

Fed QE is over for now but might not matter that much for a few months. Treasury crowding out last time showed up with a curve inversion and its a little hard to invert the curve at ZIRP, but anything below 40-50bps on the 10yr I would see as major warning sign. In fact any break below 60bps. 

https://wolfstreet.com/2020/06/18/fed-ends-qe-total-assets-drop-liquidity-injection-ends/

But other CBs still expanding so USD may slightly rise over summer? 

In the short term today we had the quad witching and with VIX in low 30s and dealers rebuilding theta it should drop next week and technicals push the market higher. I would really like to see the VIX in the teens and CTAs fully long in the next move higher. 

S&P might break through the resistance at 3214 which is the January low it was struggling with but normally you get a risk off in the summer holidays, then perhaps a bounce into election as both sides promise deficit spending, but then not make a new high while internals weaken. 

Many other assets might not rally much from here. Look for a flattening TSY curve, reasonably supported USD and peripheral assets start to flatline.



In 1929 the market bounced for about 5 months after the initial sell off, and that was under a hard money standard with no QE.

I have no idea how many of the jobs losses will be recovered in 3-6 months but it will be far less than 100%, even 60-70% would be a borderline miracle and after that, it will be the slow grind that can take years.


After the election, Trump, who I assume wins, doesnt really need to care about the stock market, he has the real economy to worry about and he cant magic up corporate margins during the worst recession since 1929. Particularly when it becomes clear the recovery is drawn-out, defaults are picking up and Schumpeter comes back from the dead like the ghost from Christmas past. 




Friday, 15 May 2020

Has my equity contribution gone to money heaven or the land of the living dead?



Low rates, QE, fiscal deficits have forced institutions up the risk curve and down the liquidity one in this cycle and two of the worst areas, in my opinion, are LBOs and blue-chip CRE yielding 4-5%, particularly in 'Global Gateway Cities' like London which are so inflated ordinary people can't afford to live in them.

We were late-cycle before Covid hit. I don't actually think Covid is that big a deal if you are not vulnerable -we don't shut the economy down every Winter due to flu. But the lock-downs initially to stop ICUs being overwhelmed with otherwise survivable working-age patients, has now transformed into a huge recession and soon a solvency and default cycle.

So I think this year corporate margins will fall heavily and then not fully recover. Additionally, the demand for CBD office space will permanently fall. The exodus of ordinary people from Gateway cities will be accelerated etc. Potentially decades of change are being catalyzed by a few weeks shutdown and its aftermath.



So what should the impact be on these two popular LP plays, LBOs and blue-chip CRE?



LBOs
LBOs have been done recently at 11-12x pro-forma EBITDA, late cyle, with more or less peak margins.

I present a few charts below from the prior blogs:







Longer run average after-tax profits as a percent of GDP are 5-7% vs 8.7% in Q3 2019, i.e. 5% after a downturn and 7% after a growth cycle. If US profits fell to the middle of that range, 6%, from 8.7% in Q3 19, profits would fall 41%, after the recovery

EDITDA is not as sensitive as after-tax profits, but let's assume a 41% profit drop equates to roughly a 30% EBITDA drop for an average LBO company. 
Let's assume a recent LBO with $1 of EBITDA was done at $11 enterprise value, financed with a $5.50 senior loan and $5.50 of equity. Then after a downturn the loan market is only willing to finance LBOs at 8.5x EBITDA. 
Using the $1 of current EBITDA, this drops to 70c. At 8.5x the 70c EBITDA the company is worth just $5.95. But the senior debt was $5.50, so the equity residual has fallen to just 45c, vs the original LP commitment of $5.50; that's over a 90% loss of equity driven by a fall in margins and a fall in mulitples. 
So you would assume this would breach loan covenants and the deal would be toast?
But here is where it gets more interesting. LBO GP's are judged, in part, on how many deals go to zero. So if a company's business has not blown up and the issue is more the amount of debt/ EBITDA, then the GP's have a huge incentive to keep a bad deal alive.
Over 5-10 years, as nominal GDP grows, most companies should see substantial revenue and therefore EBITDA growth, so by keeping a bad deal alive, even if it involves drip-feeding more equity in after a the downturn, it keeps management fees coming to the GP, a positive carry option for the GP on recovery and avoids difficult conversations with LPs, but all at the cost of the fund IRR and liquidity to the LP. 
In conclusion, with the incentives and ability of the GPs to keep bad deals alive, this looks more like a land of the living dead outlook for LBO LPs.


CRE
The situation is broadly similar in blue-chip CRE. 

With 4-5% yields recently and 7-8% post-GFC for several years. Land Securities in the UK predict a 20% drop in rent values as a function of remote working lowering overall demand levels. 
Setting aside whether I think 20% is a big enough rent drop:

Building rent: $1
4.5% yield value: $22.20
60% LTV CM: $13.30
40% equity: $8.90

Aftermath:
Building rent: $0.80
7.5% yield value: $10.70
60% LTV CM: Capital loss of $0/ -$2.60 if no Directors Guarantee
40% equity: -$2.60 if there is a Directors Guarantee

In a 50% debt, 50% equity capital structure the debt would still be under the water by $0.40.

This scenario ignores anything like skipped rents this year leading to non-payment of mortgages, rental drops of more than 20%, occupancy slumps, auctions changing hands for higher yields than 7.5% yield or Class B or C properties performing worse or going empty... 

In conclusion, Commercial Real Estate looks more like a money heaven scenario - for existing investors.