Back in March, when we detailed the ongoing catastrophic deterioration in the US retail sector, manifesting itself in empty malls, mass store closures, soaring layoffs and growing bankruptcies - demonstrated most vividly by the overnight bankruptcy of Toys "R" Us, the second largest retail bankruptcy in US history after K-Mart - we said that "just like 10 years ago, when the "big short" was putting on the RMBX trade, and to a smaller extent, its cousin the CMBX, so now too some are starting to short CMBS through the CMBX, a CDS index which tracks the values of bonds backed by various commercial properties. They are betting against securities backed by malls in weaker locations where stores could close in quick succession, triggering debt defaults."
We dubbed this retail short via CMBX the next "Big Short" trade, and others promptly followed.
In a subsequent post just a few days later, we underscored why the correct way to short the great retail collapse was not so much through stocks, but CMBX:
The trade, as we discussed before, is not so much shorting the equities where a persistent threat of a short squeeze has burned the bears on more than one occasion, but going long default risk via CMBX or otherwise shorting the CMBS complex. Based on fundamentals, the trade indeed appears justified: Sold in 2012, the mortgage bonds have a higher concentration of loans to regional malls and shopping centers than similar securities issued since the financial crisis. And because of the way CMBS are structured, the BBB- and BB rated notes are the first to suffer losses when underlying loans go belly up.
As we also noted, cracks had started to appear. As of mid-March, prices on the BBB- pool of CMBS have slumped from roughly 96 cents on the dollar in late January to 87.08 cents last week, index data compiled by Markit show.
So fast forward 6 months to today, when Goldman Sachs - a firm known to dabble with prop positions in both RMBS and CMBS in the past - itself takes aim at the CMBX trade, and in a report by Marty Young, writes that the "CMBX market doubts viability of US retail malls," which highlighting the dramatic crash in select CMBX issues we touched upon over half a year ago.
Explicitly using the term coined here first, and calling it the next "big short", Goldman writes that while 2017 has generally been a year of low volatility and tight spreads across most asset classes, the CMBX market has been a notable exception. Spreads on CMBX 6 BBB- have widened 300bp since the start of the year and now trade 385bp wide to CDX HY (Exhibit 1). More notable, the trade appears to be accelearting to the downside, and in the past six weeks alone, spreads have moved more than 100bp.
In other words, in a world in which all asset classes appears to be only going up, CMBX, and specifically the CMBX 6 BBB- tranch, has indeed emerged as this year's "Big Short."
What has driven such a significant sell-off, Goldman asks, and then provides the following answer.
A market narrative has emerged that CMBX 6 BBB- is the next “big short” of brick-and-mortar retail. The “death of retail” story is nothing new, but fresh fears have arisen this year that 2017 marks the tipping point. Following an inexplicably weak holiday retail season and a raft of store closures and bankruptcies this year, concerns are growing that the pace of deterioration has inflected higher for brick-and-mortar retailers. Although the disruption from e-commerce has been clearly visible for more than a decade, store-based retailers have been unprepared for the onslaught of online retail (“The Store of the Future,” Profiles in Innovation, August 2, 2017). The market’s increasing anxiety over regional malls and traditional anchor stores is also evident in retail stocks. Exhibit 2 shows that as equity markets have grown enthusiastic about the big names in e-commerce, they have grown only more negative on department stores. These struggling anchors in turn threaten the mall ecosystem. Markets have been acutely concerned with Sears in particular this year: Exhibit 3 shows that, with a spread level of roughly 3500bp, the CDS market is implicitly pricing a high likelihood that the company experiences distress.
For those who are unfamiliar with the basis of the trade, Goldman lays them out, as well as providing a detailed perspective on whether this "Big Short" has (much) more room to run.
First, what is it?
CMBX 6 BBB- is a synthetic, equal-weighted index of 25 CMBS mezzanine bonds that were rated BBB- at issue and issued between March and December of 2012 (CMBX 6 has AAA through BB tranches, but in this report we focus on the BBB- layer as it has been the focus of markets this year). Retail is the largest underlying property type (39%), which explains the exposure to negative retail sentiment. Office (27%) and lodging (11%) properties are second- and third-largest property types, respectively. The average loan-to-value (LTV) of the underlying collateral was approximately 64% at origination, with a debt service coverage ratio (DSCR) of approximately 1.9x. While the average deal was comprised of a pool of 64 loans, the top 10 loans account for nearly 55% of each deal on average, with the largest loan averaging approximately 10%. Most of these loans have a maturity of ten years, and thus will mature in 2022.
The tranched nature of CMBX reference entities is critical from a pricing perspective and distinguishes the product from corporate CDX. Since the most junior tranches in a CMBS deal incur losses first, there is a potential convexity to the spreads on CMBX BBB- as they need to price the possibility that these tranches could be completely wiped out. Put differently, even though CMBX 6 contains roughly 1,600 loans overall, the performance of the BBB- tranches is tied significantly to the lowest-quality mortgages in the portfolio, which is not offset by a strong performance of the aggregate portfolio. This is what makes CMBX 6 BBB- particularly vulnerable to the headwinds facing regional malls.
Exhibit 4 shows in four charts how a unique market narrative has formed around CMBX 6 BBB-. First, the spread on the BBB- tranche of CMBX series 7 ? a largely similar product comprised of CMBS issued in 2013 ? has also widened meaningful this year, but the move has been far more pronounced for the 6 series. Second, while the spreads on the synthetic index have widened hundreds of basis points since January, the spreads on the CMBS cash bond underliers have widened only a fraction of that. This deviation between cash and index could, to a degree, represent stale pricing marks on the cash bonds, given the limited trading volumes in the bond space. However, there have been enough trade prints in the sector to indicate that the pressure on spreads has been felt more acutely in index than in cash. Third, when CMBS investors have come under pressure in the past, spreads have widened at every level of the quality spectrum. This time, however, the more junior tranches have clearly underperformed, suggesting markets are pricing a scenario in which distressed assets default en masse. Fourth, the open interest in CMBX 6 BBB- has increased significantly this year, consistent with the story that this is the new "big short."
Goldman next lays out the "bear case" which as one can imagine, is substantial.
Retail malls face significant pressure from online retail, which continues to grow at roughly 15% each year, as well as fast-fashion chains and off-price retailers. Moreover, many of the anchor stores on which malls depend appear to be experiencing difficulties, and 2017 has seen the big department stores announce a host of store closures. The CDS market appears to be pricing in a high probability of distress at Sears, which would send a tremor through the mall ecosystem. And some malls have co-tenancy clauses that can amplify the impact of the department store distress by allowing other tenants to reduce their rent if an anchor closes. The slew of store closures is not limited to anchor stores, as malls are grappling with the poor performance of many national retailers. Given its significant retail exposure, CMBX faces the same headwinds that currently plague mall REITs.
The nature of the CMBX product makes it especially vulnerable to a retail downturn. First, as we noted before, CMBX is a tranched product, which means that if losses for a deal are severe enough (i.e., exceed the tranche detachment point), the recovery rates on the bonds can be 0%. By comparison, high yield corporate bond defaults usually have material recovery value. Second, the 25 deals that comprise CMBX 6 are not homogenous and the high-risk mall loans are not evenly distributed among them. If the high risk loans were spread evenly across the 25 deals, it is likely that, even if we assumed 100% losses on all of these loans, no one deal would incur sufficient losses to affect CMBX 6 BBB- in the aggregate (i.e., deal-level losses would never reach the attachment points). However, the high concentration of high-risk loans in a handful of deals threatens large losses on the product as a whole from a relatively small number of defaults.
Finally, a popular narrative that has helped drive this year’s spread widening is that Sears – commonly a tenant for many of the weaker malls in CMBX 6 – is itself at risk of imminent default. The CMBX market is worried that, if Sears defaults, it jeopardizes many of these weaker malls. For example, the Midland Mall in Midland, MI defaulted on its loan last year shortly after the Sears at that location closed. While it is difficult to tie the default directly to the Sears closure, the loss of a large tenant likely increased the pressure on the mall. Earlier this year, J.C. Penney announced plans to close its store at Midland Mall, demonstrating the potential spiral that struggling malls face upon losing a tenant like Sears. Since many of the at-risk malls in CMBX 6 share the same few large tenants such as Sears, J.C. Penney, and Macy’s, a round of store closures or a bankruptcy filing from a single retailer could do disproportionate damage to the CMBX portfolio.
To be sure, Goldman then goes through the bull case, and looks at the remittance data, which - so far - show no major signs of trouble (readers can bother their friendly Goldman sales coverage for the full report), suggesting that it is possible that the CMBX market may be getting ahead of itself. Or perhaps, like in the case of TOYS bonds, which snapped from par to 20 cents in the matter of days, what the market is underestimating is the risk of a sharp, downward inflection point as the economy, and especially US consumer, slows down further, resulting in another step wise spike in defaults.
Goldman's analyst reports as much and notes, that while the CMBX "big short" may work, it will require an inflection in performance. Here is the conclusion.
Brick-and-mortar retailers have been fighting competition from e-commerce for years. This long-running trend is visible in the rising e-commerce share of retail sales, declining same-store sales numbers, and increasing numbers of store closures. These trends, combined with the highly leveraged nature of CMBS deal structures, have fueled a bearish market narrative that has repriced the mezzanine tranches of CMBX significantly wider. This view has been particularly focused on scenarios where a subset of the lowest-quality malls generate a large number of mortgage defaults.
So far, such a deterioration in mortgage quality is not yet visible in recent vintage delinquency performance data. We find the bearish narrative persuasive, but to realize the defaults being priced by CMBX 6 BBB- will require a future deterioration in mortgage performance, which our analysis suggests would be a departure from historical predictive relationships. In our view, this “top-down” assessment highlights the critical importance of modeling the “bottom-up” credit stories. As described above, it is not hard to construct scenarios with significantly higher default losses than what we find using our narrative-free statistical analyses, and structured CMBS bonds would be highly exposed to such a collapse of the retail sector.
What Goldman is effecitvely saying is that absent a recession, or a market crash, the trade may have little widening left. Which, of course is ironic, because just several days ago, it was also Goldman that calculated that the risk of a market crash has soared to roughly 67%, as high as it was before the dot com and Global Financial Crisis crashes:
One final "hedging" observation from Goldman: in case the cautiously optimistic outlook is unwarranted, will a potential implosion in the CMBX 6 result in systemic risk? Here is Goldman's answer:
The spread widening in mezzanine CMBX tranches – and not in more senior tranches – is pointing to an expectation of high default rates on a small number of low-quality malls. If this bear case were to be realized, it would not likely cause a systemic risk event comparable to 2008, due to the low amount of mall debt relative to the amount of residential mortgage debt outstanding prior to the financial crisis. If commercial mortgage losses were to occur due to severe declines in commercial property price across all sectors – including retail, office, apartment and hotel – the impacts could be greater, given the large amount of commercial real estate exposure on US bank balance sheets. But this is not the risk scenario that CMBX markets seem to be pricing.
Of course, if a Goldman is wrong, and a terminal collapse in CMBX 6 does prompt the next systemic crisis - which of course won't be catalyzed by the losses in this segment of the Commercial Real Estate market but due to a sharper deterioration in the broader economy, all that would result in is another bailout from the Fed because as Deutsche Bank said earlier today:
"... by continually using stimulus to deal with crises and not letting
creative destruction take over, you make a subsequent crisis more likely
by passing the problem along to some other part of the global financial
system, and usually in bigger size. In a fiat currency world,
intervention and money creation is the path of least resistance. In a
Gold standard world, mining new gold was the only stable way of
increasing the money supply. we think this leaves the current global economy particularly prone to a cycle of booms, busts, heavy intervention, recovery and the cycle starting again. There is no natural point where a purge of the excesses is forced by a restriction on credit creation."