by Paul Fiorilla, Co-Managing Editor
Imagine walking into a CMBS discussion about: The role of rating agencies and credit enhancement. Transparency and the level of loan-level information provided to investors. The impact of government regulations. B-piece pricing and loan selection. The deterioration of underwriting.
Now a question: during what time period did this conversation take place?
- CMBS 1.0 (1990s through 2010).
- CMBS 2.0 (2010 through 2014).
- CMBS 3.0 (today).
- All of the above.
Of course, as every market player knows, or at least those old enough to remember the 1990s, the answer is “D,” all of the above. These issues as they relate to today’s CMBS market were the subject of a CREFC after-work seminar held Oct. 24 in Manhattan, moderated by Nomura executive director Lea Overby. The panelists were: Leland Bunch, managing director at Bank of America Merrill Lynch; Richard Parkus, managing director Seer Capital Management; Adam Smith, director at Deutsche Asset & Wealth Management; and Ken Cheng, managing director at Morningstar Credit Ratings.
New risk-retention rules will require the holders of the subordinate 5% of CMBS pools to retain the bonds for five years, which prompted a variety of reactions from the panel. Bunch called it “shocking” that there will be rules that would limit investors’ liquidity. That would particularly affect floating-rate deals, because they typically mature in five years or less. Bunch said the rule means that buyers of junior floating-rate bonds will be required to own bonds to term, calling it “preposterous.” The lower liquidity ultimately will increase the cost of capital, push up loan rates, increase rents for tenants and reduce the value of CMBS, he said.
Smith, however, said investors like the idea of issuers eating their own cooking, noting that issuers “date” loans only for a short time, while investors “marry” them for the length of a securitization. Because the rule doesn’t take effect for two years, the impact will be muted because the market will have time to adjust. “We won’t wake up in two years and say, ‘holy cow, what are we going to do?’” he said.
Overby said CMBS might ultimately follow the model of the credit-card securitization market, in which issuers retain the bottom 5% of deals.
Whatever the impact of risk-retention, it will be greatest on B-piece buyers, who own the bottom 5% of most deals. Most B-piece buyers aim to be long-term holders. Parkus noted that though there is a wide variation in strategies employed by B-piece buyers, some do in effect eat their own cooking by partnering with issuers and regularly buying the bonds from the same group of issuers.
Panelists discussed the delicate dance between issuers and B-piece buyers to avoid “kick-out” loans that are rejected from deals by the junior bondholders. If issuers are putting together a pool and find out that the B-piece buyer will reject a particular loan, the closing of the loan is likely to be delayed or put off entirely. “We spend a lot of time constantly discussing” collateral with B-piece buyers, Bunch said, “to make sure we don’t do something stupid.”
There was no dispute that underwriting took a sharp turn for the worse around mid-year. Among the symptoms are the rising amount of interest-only loans, some loans underwritten with pro-forma assumptions and a growing amount of conduit loans with construction elements. The trend coincided with a surge in the number of lenders competing for loans.
Bunch noted that specialty lenders accounted for 4-5% of CMBS during the market peak, and that has risen to about 20% today. They have taken share from insurance companies and some commercial banks, while conduits are at the same percentage. That is no accident, since reducing the commercial real estate exposure of banks was one of the intentions of banking regulations, he said.
A major difference between this cycle and the last cycle is the way the decline in standards is being treated by the rating agencies. In 2006-07 subordination levels declined as loan standards worsened. Today, the agencies have responded to the loosening of standards by increasing subordination levels. “I like the positive correlation as opposed to negative correlation,” Smith said. “If the originators make worse loans, the investors should get more credit enhancement; that’s supposed to be the trade-off.” Cheng said that higher subordination ultimately serves to reduce the profit of issuers, which could incentivize them to write better loans.
Role of B-Piece Buyers
Asked whether it was a good time to buy junior bonds at a time when loan quality was slipping, Parkus said his firm was still comfortable as long as Seer was given sufficient flexibility with the collateral pool. Though he added that “there is no question that collateral quality is headed in one direction – down.” Parkus also said that investors should differentiate between larger banks and more established players who, for the most part, have been better at holding the line on loan quality, and a subset of the new, non-bank originators who have allowed loan quality to slip significantly.
The best time to buy junior CMBS is during the early part of the cycle. “On the whole we believe we are still in an acceptable part of the credit cycle, but that view might change over the next year or two,” he said.
Role of Rating Agencies
Cheng noted that rating agency reports have become more detailed, with his agency among those that provide much more detailed information than was provided by agencies in CMBS 1.0. For example, Morningstar is one of the agencies that provides asset summaries about the Top 20 loans in pools. Agencies typically only highlighted the Top 10 loans in CMBS 1.0 deals.
Asked whether he approved of the additional information in presale reports, Smith said investors have information overload and likened researching multiple deals to triage: “It doesn’t hurt (to have more information), but the velocity of deal flow makes it difficult to get through it all.” Smith said investors are asking for standardized “best-practice” term sheets amongst the various CMBS shelves to make the investor’s job more efficient, likening the current disclosure situation to that of a “Rubik’s cube or puzzle.”
Bunch said that regulation 17g-5, which requires agencies to set up password-protected websites that detail ratings methodology, was a “huge positive” because it deterred ratings shopping and disciplined issuers to provide full information to each agency.
Panelists applauded the trend toward issues carrying split ratings. Smith gave it “two thumbs up,” saying it created discussion about why agencies give different subordination levels to the same collateral. Bunch said that issuers have to consider split ratings in terms of economics versus how each shelf wants to represent itself to the market.
Given the nature of the industry – banks write loans on commercial real estate and sell them to investors – it should be no shock that the issues involved don’t change dramatically, and probably won’t even when the market gets to CMBS 10.0 and beyond. The real question, as several panelists noted, is whether the market learns from past mistakes and takes action to not repeat them.
In some ways, lessons have been learned. As noted by Bunch, a great deal of leverage has been removed from the financial system, since investors no longer buy bonds on credit. But there still is no way to change the fact that the market is competitive and lenders will go to great lengths to win loans. “For me, the biggest concern is to reduce the fluctuations inherent in CMBS,” Overby said. “We don’t know what the next crisis will look like.”