New York | This week in The Institutional Risk Analyst, we note a couple of reader comments. First, we’ve been accused of having a fixation on the insolvent Chinese aviation conglomerate HNA. Another reader claims we have a CLO fetish in the wake of several posts on the subject. Guilty on both counts.
And it was suggested by a reader that we don’t give leveraged loans sufficient credit for the equity cushion included in the capital stack. Sure we do. But since the median debt levels of non-financial companies relative to earnings now exceed levels seen before the last financial crisis, this according to the Standard & Poor’s rating agency, we’ll stand our ground.
When it comes to risk management, watching the fastest changing datapoint on the proverbial dashboard is usually a good practice. HNA fits that description and is a modern day Icarus, an archetype for the crazy rich Asian business model so prevalent in China prior to the political crackdown by paramount leader Xi Jinping.
Then there are CLOs, our favorite asset type and the driver of the bull market in equities. The November 2018 Financial Stability report published by the Federal Reserve Board tries to spin the central bank's culpability in boosting asset prices since 2010. Notice in the table below the astronomical growth of CLOs as well as the 3x inflation rate of growth in real estate.
Why the eye-popping growth rate for leverage loans, the asset class du jour of the global speculative classes in this cycle? Low rates of course. And strangely, the figures for “real” annual price growth for commercial and residential real estate seem to be a tad low. Maybe the Board is using real inflation data to deflate the surreal real estate valuations caused by QE?
Sad to say for millions of young American families, the Bernanke-Yellen asset price inflation in commercial and residential real estate may be largely permanent, depriving a whole generation the opportunity to own a home or build wealth by investing in distressed properties. But meanwhile mortgage lending and sales volumes are in the proverbial dumper. The FRB noted last year:
“Asset valuations appear high relative to their historical ranges in several major markets, suggesting that investor appetite for risk is elevated. Spreads on high-yield corporate bonds and leveraged loans over benchmark rates are near the low ends of their ranges since the financial crisis. Equity price-to-earnings ratios have been trending up since 2012 and are generally above their median values over the past 30 years despite recent price declines. Commercial real estate (CRE) prices have been growing faster than rents for several years, and, as a result, commercial property capitalization rates relative to Treasury securities are near the bottom of their post-crisis range. While farmland values have fallen in recent years, they remain very high by historical standards. Residential real estate price-to-rent ratios have generally been rising since 2012 and are now a bit higher than estimates of their long-run trend.”
Long run trend. That’s great. Now a cynic might argue that the collapse of the CLO market in the fourth quarter and with it the larger high yield debt market was the catalyst for the latest dovish turn in the Federal Open Market Committee. We seriously doubt anyone on the FOMC actually worries about a precipitous drop in something as banal as mortgage lending, but now would be a good time to pay attention.
In particular, looking at the total issuance data from SIFMA, Q4 ’18 was a particularly bad quarter for CLOs but also saw a marked slump in issuance of all types of mortgage-backed paper as well as asset-backed securities. Could this be a sign from on high?
The US has seen a ~ 40% decrease in issuance of mortgage backed securities since the end of September 2018, certainly reason for alarm in our book. The US residential mortgage industry is looking at doing a lot less than $1.5 trillion in new mortgage origination volumes in 2018. This slump in activity is due to the volatility in the markets, which makes it impossible to accumulate and hedge new residential and commercial mortgage backed securities (CMBS) deals. Recall Q1-Q2 of 2016.
Market factors aside, the slump in residential lending volumes is more of the same, premium pricing for loans killing profitability for aggregators and banks, which are bidding points through the curve to buy larger jumbo assets for portfolio. Smaller retail lenders are dying due to high origination and funding costs. And even though rates have fallen since December, better yielding refinancing volumes are not yet returning. But hope remains: Several operators in mortgage land tell The IRA that January will be a great month.
BBT + STI: "Big Deal" or Not?
Listening to some analysts and members of the media waxing effusive about the merger of BB&T (BBT) and SunTrust (STI) last week, as in the union of two strong residential mortgage businesses, sure makes us wonder. Some 55% of BBT’s loan book is real estate loans, but the gross spread before funding and SG&A is just 440bp. At STI, real estate exposures are 40% of the loan book with a 390bp gross spread. When you subtract the cost of funding (~1.2%), commissions to the loan officer and overhead and compliance, there is not much left. Risk-adjusted returns are negative of course.
Remember that banks generally don’t care about high LTV loans with borrowers below 700 FICO, so the competition for prime mortgages is intense. The crazy high bid for mortgage assets, even as volumes drop, is why valuations for mortgage servicing rights (MSRs) are being "pulled" up to meet the execution in the secondary loan market.
The result is nosebleed 6x annual cash flow MSR multiples that cannot be validated vs, say, mainstream prepayment assumptions from the major third party advisors. And this "execution" ripples through all of the fair value MSR valuations models of the major banks. Again, special thanks and big kisses to Chairs Ben and Janet.
When investors ask if the BBT + STI combo will lead to more deals, the answer provided to the FT by no less than Rodgin Cohen at Sullivan and Cromwell pretty much sums it up. He told Robert Armstrong and Laura Noonan that the tie-up would cause “acceleration of thinking about deals, and accelerate conversations, but whether it leads to actual deals we will have to see.” He might also have said that merging two medium sized banks together at a premium to book value in an all-stock deal is hardly reason for great excitement.
A few years back, when we'd drop off documents at Sullivan & Cromwell on the way home from the Federal Reserve Bank of New York, banks did not trade much above book value. The reason for this, quite simply, is that well run banks have low double digit equity returns, but that it about it.
At present BBT’s cost of funds is about 1.4% vs 1.5% for Peer Group One, which includes all 120 banks in the US above $10 billion in assets. BBT’s gross spread on total loans and leases was just 4.69% at the end of Q3 2018, according to the FFIEC. This is just below the peer average.
In other words, BBT and STI are middle of the fairway performers, consistent earners and with strong capital, but not great sources of alpha. Together these banks are not worth much more than the current 1.4x book value. Net loan and lease growth for BBT over the past five years is modest at best, begging the question as to why certain analysts are getting so excited about “the biggest bank deal since 2008.” To be fair, the larger banks are even worse in terms of risk adjusted returns on real estate loans.
JPMorgan Chase (JPM), has 42% of total loans in real estate loans for a gross spread of 384bp. That includes both residential and commercial exposures. Bank of America (BAC) has a third of its loan book in real estate loans and is working for a whole 386bp gross, before funding costs or sales costs or overhead expenses. Layer on a charge for the reputation risk element of facing consumers and the equity returns on residential mortgages quickly go negative. H/T to Liz Warren and Kamala Harris.
We might have been tempted to own BBT prior to the transaction, but adding STI to the mix does not make us want to own the post-merger bank. Much rather add to the position in U.S. Bancorp (USB), which has lower funding costs and a more diverse revenue mix. And even after the market volatility seen that the end of 2018, the financials as a group are not particularly cheap. Just remember that in a ranking of large US banks based upon equity returns, USB is the only bank in the top 10 by assets that makes that illustrious list of exemplars in terms of equity returns.
Of course, for all of you who still think that rising rates are good for banks, the fact that the 10-year Treasury note has rallied since last Thanksgiving is probably not welcome news. And the bond market rally is unlikely to slow the "normalization" of bank interest expense, which has seen the cost of funds for the industry rise 75bp in 2018 by our calculations. Indeed, after almost touching 5% around turkey day, the 30-year fixed rate mortgage has fallen below 4.5% and seems to be following the surging T-note lower in yield.
Since early January, high yield spreads have taken their queue from the equity markets and rallied more than a point in yield. Option-adjusted spreads have essentially collapsed. Indeed, just about every bank asset class we can think of from loans to securities and mortgage servicing rights are surging higher, buoyed by "the new confidence." Echoing ECB head Mario Draghi, the FOMC will do whatever it takes to keep the world safe for heavily indebted corporations, Italian banks and anyone else with the slightest potential for causing a deflationary market event. Have a great week.
The Vice President’s Men
Seymour M. Hersh
The London Review of Books, 24 January 2019
Jim Grant on the Bond Market’s 35-Year Bull Run Barron's, February 1, 2019 https://www.barrons.com/articles/sizing-up-the-bond-market-51549050648
The Fed's Killing Floor Zero Hedge, 2/6/19 https://www.zerohedge.com/news/2019-02-06/feds-killing-floor