An article in the Financial Times by Tracy Alloway gives yet another sighting that bond investors are getting a bit frantic in their hunt for yield. The piece has the eyepopping title, Yield-hungry investors snap up US homeless bond. It uses recent deals in the CMBS (commercial mortgage backed securities) market as a proxy for bond investors’ QE-driven hunt for more return.
And the homeless part isn’t an exaggeration:
One of the more eye-catching examples is a bond sold by Citigroup which is backed by 137 commercial mortgages. The bond includes a loan to 127 West 25th Street, a homeless shelter whose location in the fashionable Chelsea neighbourhood of Manhattan has raised the ire of some of the area’s residents. The bond was sold last year but has not previously been reported.
Now as sensational as this sounds, this data point is hardly damning in isolation. Who is the lessor? New York City or a private body? If New York City, and the lease is long enough and has decent price escalations built in, this could actually a good if unconventional credit. And having 5-10% of a deal in unusual leases isn’t imprudent if the leases have stable enough tenants and/or compensating factors (like the yield on them really does compensate for the risk). But this is still a big departure from the usual constituent elements of CMBS deals, whose staples are conventional commercial properties, like high quality office buildings and malls. And the homeless shelter is far from the only exotic asset included in these transactions. One which is clearly high risk is the Kalahari Resort and Convention Center, “a chain of African-themed waterparks.”
And the article makes clear that there are broader signs of frothiness:
Pro forma underwriting, where originators estimate the future cash flows of a property on optimistic projections rather than current rates, has crept back into the market, according to some commercial mortgage brokers who connect property developers with lenders.
“Different folks have different perspectives on what they believe ‘pro forma’ is,” said Eric Thompson, senior managing director at Kroll. “There is a clear line though – and if originators are giving credit to rental revenue that is not in place and/or above market rates it is pro forma, and we have seen some of that.”
Interest-only loans, another hallmark of pre-crisis lending practices, have also returned.
Another indicator is a close-to-cheerleading article a few days ago in the Economist about the return of securitization. Never mind that securitization was used to eliminate the need for banks to hold equity against loans, and over time, lending decisions based on local market knowledge and character assessment were replaced by model-based processes. The latter did not perform very well. Yet the article blandly parrots official PR that the problems with securitization have been remedied:
Why are regulators so keen on the very product that nearly blew up the global economy just five years ago? In a nutshell, policymakers want to get more credit flowing to the economy, and are happy to rehabilitate once-suspect financial practices to get there. Some plausibly argue it was the stuff that was put into the vehicles (ie, dodgy mortgages) that was toxic, not securitisation itself. This revisionist strand of financial history emphasises that packaged bundles of debt which steered clear of American housing performed well, particularly in Europe….
This is in large part because regulators want banks to be less risky, by increasing the ratio of equity to loans. As banks are reluctant to raise capital, they need to shed assets. This is where securitisation helps: by bundling up the loans on their books (which form part of their assets) and selling them to outside investors, such as asset managers or insurance firms, banks can both slim their balance-sheet and improve capital ratios.
Securitisation “airlifts assets off the balance-sheets of banks, freeing up capital, and drops them onto the balance-sheets of real-money investors,” in Mr Haldane’s words…
One improvement is that those involved in creating securitised products will have to retain some of the risk linked to the original loan, thus keeping “skin in the game”. The idea is to nip in the bud any temptation to adopt the slapdash underwriting practices that became a feature of America’s mortgage market in the run-up to the financial crisis. Another tightening of the rules makes “re-securitisations”, where income from securitised products was itself securitised, more difficult to pull off.
Let’s consider how this really works. Banks have been repeatedly required by investors to rescue securitizations they’d sponsored. In other words, they really didn’t transfer the risk to investors. For instance, consider the history in credit card conduits. As we wrote in ECONNED:
And readers who followed the press in the early phases of the crisis will also recall how banks were on the hook for the supposedly off balance sheet structured investment vehicles (recall Citigroup board member Bob Rubin claiming he was unaware of the fact that they contained a “liquidity put”.)
Nor should anyone put much faith in “skin in the game”. The amounts are too small in economic terms to have much impact, and any losses are down the road and hurt the institution, rather than the responsible individuals. Clawbacks of compensation for the teams who worked on these deals would be far more effective. The article also claims that investors are wiser about CDOs. History shows that memories fade with remarkable speed. The 1990s subprime market, like its successor in the 2000s, depended on CDOs to sell the riskiest tranches of subprime bonds. That market blew up at the turn of the century. CDOs made a comeback in a small way in 2003, and had roared back to life by 2005, because supposedly new better structures and asset selection had fixed the problems with the old CDOs. Regulators should have prohibited resecuritizations entirely (this was part of a very well-thought out FDIC proposal that the sell-side fought hard). The fact that they’ve left an opening will allow the industry to push it wider.
And an aside early in the piece seriously undermines the author’s credibility:
Excluding residential mortgages, where the American market is skewed by the participation of federal agencies, the amount of bundled-up securities globally is showing a steady rise (see chart).
The reason the Federal agencies are so deeply involved in the US mortgage market is that US mortgage investors see securitization reforms as inadequate and are not getting back in the pool. The Federal agencies are filing a huge gap, not pushing private capital out, as the Economist incorrectly implies. And part of the failure of securitization in the US is the inability of investors (when they could not, as with credit cards, make a stink and force the banks to deal with unexpectedly high default levels) to get recourse in court for a whole raft of abuses: misrepresentation of the quality of the deal, the trustees’ failure to replace servicers who refused to put back bad loans to originators, and numerous types of servicing abuses, which hurt borrowers as well as investors. The failures here are legion, yet the Economist brushes by them.
Finally, a comment on the FT article by one “William J. Harrington (Former Moody’s SVP & Derivatives Analyst)” highlights another not-widely-recognized risk, that of defaults on the swaps in these deals (and the swaps are senior in these structures):
Commenters are correct that taxpayers are on the hook. Most CMBS, indeed most U.S. ABS, are exposed to counterparty risk, in addition to asset pool risk. Issues of CMBS and other ABS hold zero reserves against failure of a counterparty to an interest rate swap, instead relying on unenforceable “flip clauses” to pretend that an insolvent counterparty cannot claim funds earmarked for investors. Similarly, all rating agencies, including those quoted above, assign zero probability to the risk that CMBS investors will lose if a counterparty becomes insolvent.
Bank bail-outs stemmed CMBS and ABS losses in 2008 by preventing a major counterparty from becoming insolvent (Lehman provided few interest rate swaps to ABS issuers.) Only future bail-outs will validate issuers in betting that zero reserves will be more than ample to protect CMBS investors against counterparty exposure.
In fairness, the fact that investors are feeling their oats more than looks prudent does not mean a crash is nigh. Gillian Tett started expressing muted concerns about CDOs in 2005, for instance, just as the toxic phase was starting, years before the ugly denouement. But the hunger for risk-taking and the regulators’ confidence that meager reforms were adequate are not the sort of thing that one wants to see so soon after a global blow-up.