Should you sell bonds/hybrids to buy ASX listed RMBS?

Switching out of senior bonds and hybrids to invest in Australian residential mortgage-backed securities (RMBS) while house prices are falling---and likely to remain weak for years---is one of the more surprising trades I have seen of late. It is right up there with selling major bank hybrids on the ASX to buy the new WBCPH or CBAPG issues that offer returns that are 0.2 per cent to 0.6 per cent below the hybrids folks have been dumping. For the patient, this irrational behaviour can furnish tremendous opportunity.

We were a happy buyer of AAA rated RMBS when house prices were booming at more than 10 per cent annually, the Reserve Bank of Australia was slashing interest rates (reducing mortgage default risks), and there was scant new supply. In the last few months we have completely exited the sector, taking profits on circa $200 million of RMBS because every single one of these tail-winds have reversed.

Across the five largest capital cities house prices have been gradually declining since October 2017 with values dropping by 2 per cent thus far. Bond markets are pricing in higher, not lower, RBA cash rates, with a minimum of 2 to 3 hikes expected over the next 3 years. In contrast to extremely benign corporate bond default rates, default rates on home loans have climbed from their post-crisis trough of 0.6 per cent in 2015 to around 0.75 per cent today (see these RBA charts) and will trend higher as the cash rate increases (RMBS represents a bundle of home loans). 

And the supply of RMBS issues has exploded (see here) with almost $40 billion sold in 2017, double the volume in 2016, and almost back to the circa $50 billion annual issuance levels seen before the 2008 crisis. One canary in the supply-side coal-mine is the volume of non-bank issuers, which in 2017 was, according to the RBA, “by our count 17---the highest ever”. Non-banks like Latrobe, Liberty, Pepper, Bluestone and FirstMac tend to issue RMBS portfolios with significantly higher default rates than the banks and pay correspondingly superior interest rates.

Yet it was also these securities that suffered the worst illiquidity and mark-to-market losses during the crisis. As a rule we have historically refused to invest in any RMBS issued by non-banks because of the greater agency conflicts. When a bank sells you a portfolio of its home loans via an RMBS deal, you know that it keeps the vast majority of its residential mortgages on its balance-sheet. It will hold them to maturity and experience broadly the same default risks as its RMBS issue.

A non-bank, by way of contrast, does not hold loans to maturity and originates them to sell to third-party investors, which means there is less alignment of interest and a greater chance that the non-bank applies looser credit standards when originating new loans in the name of maximising the volume they can off-load to investors. While this can be mitigated somewhat by the non-bank retaining a first-loss equity exposure to the RMBS deal, my experience is that non-bank credit risks are significantly higher than mainstream Aussie banks.

Some investors have asked me about a new listed investment called the Gryphon Capital Income Trust, which expects to be 85 per cent invested in RMBS and 13 per cent invested in similar asset-backed securities (ABS) with 2 per cent in cash.

While the product disclosure statement says Gryphon invests in "secured" RMBS and ABS that sit below government bonds and above senior unsecured bonds (and hybrids) in the capital structure, this does not tell us much about its actual credit risk.

An RMBS and ABS deal is broken up into many different tranches, or subsidiary bonds, ranking from the super-safe AAA rated tranches down to extremely illiquid and risky BB or B rated sub-investment grade, or “high yield” (aka “junk”), tranches.

Gryphon says it will primarily invest in the riskier and less liquid A, BBB, BB, B and unrated tranches. On the basis of these ratings, this is equivalent to investing in subordinated bonds, hybrids and even riskier high-yield securities.

The real tell however is in Gryphon’s target return of the cash rate plus 3.5 per cent. On the latest Latitude (ABS) and LaTrobe (RMBS) deals launched in the last two weeks, you would have to be 100 per cent invested in the junior-ranking A, BBB, BB and B tranches to deliver investors a return of 3.5 per cent above cash after fees. 

Standard & Poor’s says that a 10 per cent drop in Australian house prices will result in LaTrobe’s A rated tranche falling to BBB-, its BBB tranche being downgraded to a sub-investment grade (or junk) BB rating, and the BB tranche declining to a very low single B rating.

By way of comparison, the major banks’ listed BB+ rated hybrids are currently paying up to 2.75 per cent above the cash rate excluding all franking (eg, CBAPD) or 3.70 per cent above cash including franking for the 92 per cent of taxpayers that can still use these credits if Bill Shorten's tax vision is realised.

While we invest in both RMBS and hybrids in some mandates/portfolios, I prefer the superior liquidity and choice of the 30 to 40 listed bank hybrids than poorly rated tranches in illiquid RMBS that will suffer if house prices fall further. And that is especially the case after hybrid spreads have blown out about 0.90 percentage points (or 90 basis points) on the back of circa $3 billion of new supply care of the WBCPH and CBAPG issues, and the Bill Shorten-related noise on cash rebates (franking itself remains rock-solid).

In the capital stack right now you can get about 1 per cent above the 1.5 per cent RBA cash rate from a term deposit, 1.45 per cent above cash from the major banks’ 5 year senior bonds, 2 per cent above cash via a major bank subordinated bond, and 2.4 per cent to 2.75 per cent above cash from hybrids assuming zero franking for people who pay no tax, or 3.6 per cent to 3.8 per cent above cash for taxpayers.

Spreads on the major banks’ senior bonds have recently jumped from 0.75 per cent above the bank bill swap rate to almost 0.95 per cent today. At the same time, the bank bill rate has leapt from 0.25 per cent above cash to around 0.50 per cent on the back of movements in US short-term rates, which provides for an unusually attractive total return for these liquid and AA- rated securities.

The increase in major bank senior spreads has been partly driven by at least three bank balance-sheets selling these assets to build cash to improve their liquidity coverage ratios (LCRs) as they approach their March 31 half-year balance dates. This has been compounded by Japanese investors’ full-year balance dates also arriving on March 31, which has temporarily removed an important bid. A global softening in spreads has provided a final tail-wind for spread widening.

With markets welcoming a relatively dovish Fed rate hike, we should see some stability in spreads post-Easter as global default rates remain benign and growth continues to intensify with the help of stimulatory monetary and fiscal policy settings that will stay in place for as long as wage and inflation growth stays below the central banks' targets.