In the AFR today I write that for those willing to fade the extreme moves in financial markets during March, April offered up an immensely profitable case of so-called “regression to the mean”. This concept of mean-reversion was first characterised by the British polymath, Sir Francis Galton, in his late 19th century study of genetics. Click on that link to read the full column or AFR subs can click here. Excerpt enclosed:
In financial or bank-issued bond markets where the cost of capital is ultimately determined by central banks, we have observed a predictable and yet nonetheless stunning retracement in credit spreads.
When the world was gripped by extreme fear in March, and investors were universally rushing for the exits, the spreads on the major banks’ bonds leapt with unprecedented speed to extreme, and in some instances, record levels. Five-year senior bond spreads jumped from 69 basis points above the bank bill swap rate (BBSW) to as wide as 175 basis points over, broadly in line with the levels last touched during the darkest days of the global financial crisis (GFC).
The major banks’ five-year subordinated bond spreads smashed through their GFC “wides”, increasing from 160 basis points over BBSW to as high as 400 basis points (the GFC peak was around 300 basis points). And in the hybrid market, spread moves offered extraordinary opportunities as the five year major bank curve exploded from 270 basis points over BBSW in January to as high as 840 basis points in March, way above the GFC peak of circa 590 basis points.
Folks like to talk about being contrarian, but few have the courage of their convictions. Many were exiting these markets at the worst possible time. I am lucky to have some genuinely iconoclastic clients that are prepared to bet against hysteria and exploit the ineluctable regression to the mean in financial spreads.
Three in particular—call them the Jedi, the Adonis, and Croaky—pumped $100 million into the major banks’ senior bonds and hybrids during the toughest days in March (I net bought about $800 million of bonds in March). In the ASX hybrid market, which proved to be very liquid in March with up to $120 million trading on individual days, these investors earned capital gains north of 15 percentage points (pre coupons) in just a few weeks. And there is still loads of upside left with investment-grade major bank hybrid spreads some 150 to 200 basis points wider than they should be.
With perfect foresight, you could have very successfully deployed similar mean-reversion strategies in the big shocks in 2009, 2012, 2016, and 2018. The essential driver of this regression is the fact central banks are tasked with controlling borrowing costs. When these interest rates spike to unseemly levels, monetary policy’s mission swiftly adjusts to bringing them back down to earth. The alternative is that the cost of capital sky-rockets at the worst possible time, severely amplifying the recession.
While the mean-reversion in the second half of March and April was a high probability event ex ante, the initial blow-out was much more speculative. In mid-January the Chinese informed everyone that there was no evidence of human-to-human transmission of COVID-19, which the World Health Organisation faithfully reaffirmed. While this proved to be the mother-of-all head-fakes, a reasonable person would have assumed the WHO’s analysis was credible.
Even allowing for the coronavirus’s highly infectious reproduction rate, which was eventually revealed, we now have clear counterfactuals that containment after an initial outbreak is absolutely possible. In countries like Taiwan, Hong Kong, New Zealand, and Australia, early outbreaks have been effectively eliminated following assertive lockdowns, border controls, and comprehensive testing and tracing.
It was not, therefore, a foregone conclusion in January that COVID-19 was going to become a pandemic. Had the Chinese not waited until January 23 to shut-down Wuhan, not left their international borders wide open during the subsequent Chinese New Year holiday, and not lied to the world, the outbreak could have been prevented.
To conclude COVID-19 was going global, one had to carefully monitor the flow of infection data. We built real-time data-tracking systems for this purpose, and it was pretty obvious in the second half of February that containment failures would propagate a pandemic. At this point, we argued that the patent inability of financial markets to properly price and trade a pandemic would precipitate their outright failure and an ensuing liquidity and solvency crisis. But the liquidity/solvency crises could have been pre-emptively cauterised in late February and early March through actions by central banks and treasuries to provide unconditional liquidity and asset pricing support via aggressive quantitative easing (QE) policies.
We now know this to be true because that is precisely what QE around the world achieved over late March and April. That is, an enormous improvement in the liquidity of bond markets coupled with a dramatic decline in the cost of capital of high-grade credit. Our mistake was to assume policymaking perfection and think that this solution would be forthcoming in early March: it took the savage price action in the first half of that month to convince central banks and treasuries to act, which they eventually did ex post facto.
This is not to take away from the exceptional agility and resolve our monetary policy mavens and politicians have displayed. Once the situation was synthesised, the Reserve Bank of Australia was magnificent, unleashing its QE bazookas on March 19. The prime minister, Scott Morrison, has also risen to the occasion, despite consistent efforts by States governments to sheet home blame for their own failings, as they did during January’s devastating bushfires. We saw a similar playbook in this tragedy, with New South Wales and Victoria pre-announcing policy decisions ahead of National Cabinet meetings in an effort to make the prime minister look silly.
This time around ScoMo saw the dummy coming and adjusted his defence. As I have explained before, one of his strengths is his unusual (for a politician) propensity for preternaturally underpromising and overdelivering. The PM has done this impressively with COVID-19. The shock-and-awe of ScoMo’s conservative “six month hibernation” plan helped to crush Australia’s infection curve more effectively than most countries globally save for New Zealand and Hong Kong.
In March we advised the prime minister that he would need to pivot away from this six month containment policy towards a much more rapid than expected exit one to two months after the lockdown’s initiation. This was partly because our proprietary forecasts for COVID-19’s trajectory signalled that the peak in the Australian, US and European curves would pass in early April, months before most epidemiologists’ estimates. This has been borne out by the data since with the likes of Australia, the US, UK, Italy, Spain and France all moving through their peaks in the first half of the month.
And while the prime minister did not initially embrace our assessment, he has promptly pivoted to a graduated exit in May miles ahead of schedule as the data validated our projections. This has important consequences for fiscal policy and Australia’s AAA credit rating. At around 10 per cent of GDP, the original fiscal policy package was designed to thwart the full-blown depression that would follow from shutting-down the economy for half a year.
If we can normalise activity much more quickly, the PM and Treasurer Josh Frydenberg can prudently pare back this stimulus, which would (once again) save Australia’s AAA rating, sparing borrowers unnecessary interest rate increases. ScoMo did this in 2017 and 2018 by bringing the budget back to balance years ahead of schedule. And I reckon he and his successor will do it again.
Similar prudence has been evidenced with the major banks’ cautious capital management plans. We expected one major bank to defer its dividends, which ANZ has done, and the others to slash them. While this loss of income hurts shareholders, it appropriately protects creditors and ultimately the banks’ cost of funding. While the banks’ share prices have reacted poorly to declining dividends, the credit spreads on their bonds have compressed materially as investors reward their desire to build world-beating capital buffers. It is effectively a transfer of wealth from borrowers (equity) to creditors (debt).
If you want to watch or listen to a Youtube podcast I did this week on our macro outlook, COVID-19, housing and much more, click on this link.
Never miss an update
Stay up to date with my content by hitting the 'follow' button below and you'll be notified every time I post a wire. Not already a Livewire member? Sign up today to get free access to investment ideas and strategies from Australia's leading investors.