Transcription

WHARTON REAL ESTATE REVIEW SPRING 2012The Past, Present, and Future of CMBSSAM CHANDANIn the aftermath of the savings and loan crisis, commercial mortgage securitization emerged as the dominant sourceof new financing by the property market's peak in 2007. Its roost atop the lending hierarchy was short-lived, however,as the next year's financial collapse upended securitization across a broad range of asset classes. Issuance ofcommercial mortgage-backed securities (CMBS) fell from 229 billion in 2007 to 12 billion in 2008, and just 3billion in 2009 (Figure 1). For all practical purposes, the CMBS market had ceased to function. Interventions designedto trigger new CMBS activity during this period, including the non-legacy provisions of the Term Asset-BackedSecurities Loan Facility (TALF), were largely unsuccessful.Following three years of dormancy during which the performance of legacy CMBS loans deteriorated sharply, a spateof new issues in the first half of 2011 raised expectations of a revivified conduit. In spite of strong demand for thesedeals, a macro finance environment that exacerbated investor risk aversion and roiled bond markets negated the earlymomentum. Spread over swaps for AAA-rated CMBS widened from a low of 170 basis points in early 2011 to morethan 300 basis points in the fourth quarter, complicating loan pricing. Issuance for the year ultimately reached 33billion, prompting some conduit originators to shutter.The ratcheting back of activity has allowed investors to revisit persistent structural weaknesses in the securitizationmarket, such as potential conflicts in issuers' selection of ratings agencies. Specific events, such as the well-publicizedcollapse of an August 2011 deal, have resuscitated questions linked to other principal-agent conflicts as well.Balancing these issues against a recalcitrant shortfall in secondary markets' access to financing for new transactionsand maturing debt, CMBS activity is projected to resume its recovery during 2012. The long-term outlook is qualified,however, as industry-led and regulatory reforms compete to shape the next generation of securitization.CMBS PastUntil the savings and loan crisis, traditional sources of financing dominated the commercial real estate industry. Themass failure of thrifts and the broader disengagement in real estate lending saw alternative forms of financing grow inimportance. Listed real estate investment trusts (REITs), which had been a peripheral feature of the market duringthe early 1980s, increased their aggregate market capitalization more than fourteen-fold between 1987 and 1997.During the same period, the establishment of the Resolution Trust Corporation (RTC) through the FinancialInstitutions Reform, Recovery, and Enforcement Act of 1989 (FIRREA) gave rise to the modern CMBS market.Bonds backed by commercial mortgages were not an entirely new phenomenon. In fact, bonds were a measurablesource of financing as early as the 1920s and 1930s. At that time, securitization structures were much simpler, withbonds generally backed by a single loan and property. Returning to the modern era, the first collateralized mortgageobligation (CMO) was issued in 1984 by the Penn Mutual. In spite of beneficial developments, including theintroduction of the Real Estate Mortgage Investment Conduit (REMIC) with the Tax Reform Act of 1986, issuancewas tepid in the following years, with cumulative CMBS volume only passing 10 billion in 1991.Following the establishment of the RTC and its first rated issue in January 1992, CMBS volume swelled. In contrastwith the market's early experience with commercial mortgage bonds, the RTC employed the modern CMBS structure

to pool large numbers of mortgages into a single issue. This strategy was necessitated by the sheer volume ofcommercial mortgages on the balance sheets of the failed thrifts. Demand for RTC CMBS issuance was tepid at first.Measures taken to improve transparency, including the creation of the Performance Portfolio Report (PPR),contributed to market acceptance of twenty-seven deals with an aggregate balance of more than 17 billion includingcommercial and multifamily loans. By 1997, the cumulative losses to bond holders amounted to only 3 percent of theoriginal mortgage balances, a fraction of the rating agency credit reserves of 26 percent.While the RTC incubated CMBS through a period of extreme distress in property markets, the securitization vehicleoutlived its initial raison d'être. By the late 1990s, the CMBS market had shifted from a tool for managing seasonedloans originated and held on distressed balance sheets to a source of financing for new mortgages originatedspecifically for securitization. Adding to the attractiveness of securitization, FIRREA imposed new capital rules thatrequired savings institutions to hold risk-based capital equal to 7.2 percent (initially 6.4 percent; 8.0 percentbeginning December 31, 1992) of their risk-weighted assets. For some institutions, new commercial mortgage lendingwas also constrained by discrete concentration limits.Figure 1: CMBS issuanceSource: Chandan Economics.With the costs of portfolio lending rising, the institutional structures of CMBS markets matured. Bond ratings openedinvestment to a wider range of passive investors and secondary market liquidity increased. By 2001, CMBS displacedlife company loans as the second largest pool of commercial mortgages outstanding (Figure 2). During 2007, the netincrease in CMBS balances surpassed banks (Figure 3). The latter had been under increasing regulatory andsupervisory pressure since at least early 2006, when the major bank regulators formally expressed "that someinstitutions have high and increasing concentrations of commercial real estate loans on their balance sheets," addingthat they "are concerned that these concentrations may make the institutions more vulnerable to cyclical commercialreal estate markets."Figure 2: Commercial mortgages outstanding by lender group, excluding agency multifamily and construction loans(in trillions)

Source: Federal Reserve.Figure 3: Net change in commercial mortgages outstanding by lender group excluding agency multifamily andconstruction loans (in trillions).Source: Federal Reserve.

To the extent that deepening credit markets and low interest rates were amongst the complex set of factors thatcontributed to rising property values during this period, passive investors' diminished perceptions of risk enhancedthe competitive position of CMBS versus balance sheet financing. Investors' required yields fell, reflected in narrowerbond spreads that allowed conduit lenders to compete more aggressively with banks (Figure 4). Even as debt yieldsdeclined, a larger share of each deal was assigned to the senior or super-senior tranche. For example, AAA creditenhancement generally exceeded 30 percent in 1996 and 1997 but had fallen to an average of just 12.1 percent by2007 (Figure 5). By way of context, estimates of the losses on the 2007 CMBS vintage have been projected to exceed20 percent.Figure 4: CMBS AAA spreads over swaps (basis points).Source: Trepp.An endogenous relationship between property prices, investor assessments of risk, and declining subordinationrequirements during this period showed little sensitivity to the cyclical default and loss characteristics of commercialmortgages. A self-reinforcing cycle of low-cost credit, higher leverage, and rising prices limits the potential for selfcorrection until some unsustainable level is reached, triggering a crisis during which property values fall below theirindebtedness.In the case of the CMBS market, the market structure itself may exacerbate pro-cyclical lending. Ratings shopping,wherein market structures may allow issuers to select the ratings agencies that offer the most favorable creditassessments, contributed to a decline in bond and loan quality as well. In their working paper on CMBSsubordination, Richard Stanton and Nancy Wallace of the Haas School of Business attribute changes in subordinationto regulatory-capital arbitrage, finding that "the difference between AAA CMBS yields and AAA corporate bond yieldsfell significantly in the years after 2002, when risk-based capital requirements for highly rated CMBS were loweredgreatly. No comparable drop in relative yields occurred for lower-rated bonds."Figure 5: Average and range of AAA credit enhancement level by vintage.

Source: S&P.Pre-crisis research attributes some of the incentives for securitization to arbitrage or reputational considerations,finding that loans retained on bank balance sheets experienced higher rates of default than mortgages that weretransferred to securitization. In the denouement of the market peak, delinquency and default rates on commercialmortgages increased rapidly. The performance of CMBS mortgages deteriorated further than for other lender groups,which may reflect changes in the conduit origination path and the competitive position of conduit originators andportfolio lenders in the final years of the run-up. Differences in CMBS loan performance across originator types alsoappear to have been significant determinants of variation in loan performance, reinforcing that incentives and moralhazard are potential and realized features of the prevailing structure of securitization markets.CMBS Present And FutureFollowing a protracted period of illiquidity and retarded price discovery, investment, asset prices, and creditconditions began to improve in late 2009. That improvement was circumscribed, however, with gains concentratedamong high quality assets and borrowers in a small set of historically very active investment markets. Withpreferential access to capital, listed REITs became the dominant (net) acquirers of assets in 2010 and 2011. Forcompeting bidders with a greater reliance on secured financing, life companies, foreign bank lenders, and nationaland large regional banks have been the primary sources of new credit. In 2011, life companies were the only majorgroup to register a net increase in commercial real estate lending. Still, spreads on loans in the office and apartmentsectors narrowed for the most coveted assets, falling below historic norms by mid-2011.As CMBS lending has resumed under a "CMBS 2.0" moniker, its impact has been observable across a much broadergeographic area. In primary markets, CMBS in late 2010 and early 2011 accounted for less than 10 percent of loans.Even at its relatively low volume, it accounted for approximately 15 percent and 25 percent of issuance in secondaryand tertiary markets, respectively. Reliant on well-diversified collateral, conduit lenders have necessarily grown moreactive outside of cardinal markets, in locations where the lender landscape is more thinly populated, where lendingcompetition is more subdued, and where higher spreads are more supportable. Retail and industrial properties havebeen the primary beneficiaries of new CMBS activity (Table I). Multifamily lending, on the other hand, has accountedfor a de minimis share of issuance on account of more favorable terms available through agency financing.

Table I: Original balance characteristics of CMBS loans originated and securitized in 2011 ( millions)Excludes single-borrower deals and loans accounting for more than 30 percent of total deal balance.Source: Chandan Economics.Constraints on many regional and community banks' capacity to extend credit in support of smaller markets'commercial property sales and refinancing needs are likely to persist for some time. Particularly in cases where thebank lender has significant exposure to legacy construction and development lending and in cases where themanagement of legacy distress has proven unwieldy, regulatory and supervisory pressure may require a long-termdrawdown of exposure to the commercial property sector. Given a paucity of alternative credit sources in thesemarkets, liquidity has been significantly enhanced as conduit lending has increased. The immediate value of CMBS inmeeting the market's financing needs will increase as more legacy debt matures over the next several years.Almost as soon as CMBS issuance resumed, the less conservative geographic and property mix began to raisequestions about the quality of assets being bundled into new issues and concentrations in the largest loans. Stressedmeasures of loan risk have risen. As compared to pre-crisis issuance, new deals are more disciplined. Consensusbenchmarks for a target level of discipline are elusive, so the quality debate continues. For market skeptics, theabsence of a comprehensive structural overhaul of CMBS means the current bias in favor of conservativeunderwriting will ultimately give way to greater risk-taking as the market normalizes. While the mix of loans andunderwriting standards embedded in forthcoming deals may weed out more risk-averse investors, there are otherreasons to be cautious in welcoming a resurgence of CMBS activity. Aside from the record-high volume of loans inspecial servicing, many of the structural issues that were material contributors to the CMBS market's crisis-periodcollapse remain unaddressed or poorly understood.The Commercial Real Estate Finance Council (CREFC) has made progress in addressing investors' desire for a moretransparent and well-functioning market. The recently released CMBS 2.0 standards are evidence of that progress.Among the other provisions, new guidelines address lenders' representations and warranties to investors regarding anissue's loans and the due diligence performed on properties and borrowers. The guidelines also provide modelremediation language, outline underwriting principles intended to minimize the risk of loan non-performance, andfurther standardize issues' Annex A files.Conclusion

The industry's progress in enhancing its capacity for self-regulation is laudable. However, the empirical findingsregarding incentives raise serious questions that extend beyond the achievement of CMBS 2.0. In the near- tomedium-term, uncertainty related to methodological standardization across ratings agencies, the performance of thelegacy market under increased stress from maturing debt, and risk retention requirements specified in Section 941 ofthe Dodd-Frank Wall Street Reform and Consumer Protection Act have fueled a divergence of expectations about dealflow. Opponents of across-the-board risk-retention requirements for CMBS argue that it will unnecessarily raise costsand inhibit issuance. In May 2011 testimony before the Senate Banking Committee, the CREFC pointed to the initialimprovements in CMBS volume and cited the "securitization risk retention framework mandated by Dodd-Frank [as]the biggest threat to sustaining that recovery." There is certainly a danger that some elements of the implementationwill have deleterious consequences unanticipated by the Act's framers.Even if some implementations of risk retention will adversely impact issuance, the motivation properly reflect thatCMBS pools have performed poorly during this cycle and that the market structure is partly indictable in thisoutcome. As noted in a recent Federal Reserve Board of Governors report, "the financial crisis has highlighted severalways in which the incentives of participants in securitization markets may have been misaligned with incentives onewould expect to find in a well-functioning market." The elevated rates of CMBS delinquency and default areconsistent with structural weaknesses in the securitization market that are not replicated elsewhere to the samedegree. The impact of market failure is not contained; the externalities from CMBS losses are felt across the propertymarket.One example of these spillovers presents a challenge for the nation's banks, where the incentives to project long-termloan performance at origination are different from those for conduit lenders. Rising competition from the conduit canultimately undermine loan quality among its competitors, including the regulated banks. CMBS may be underwrittenmore carefully now than a few years ago, but this cyclical focus on risk is not a substitute for measures that will ensurethe long-term health and sustainability of the industry. The structural assessment of CMBS that addresses the rangeof incentive conflicts with practical and implementable remedies is incomplete. In the best case, industry-led effortswill intensify, limiting or precluding inflexible regulatory regimes and encouraging rather than impeding robustsecuritization activity. In the worst case, inattentiveness to incentive conflicts will allow the current bias in favor ofconservative underwriting to cede to undue risk-taking later in the cycle. In the immediate future, demand for newissuance will be evidence of sufficient lessons learned and the cycle will threaten to repeat itself again.Copyright 2012 Sam Chandan.

commercial mortgage-backed securities (CMBS) fell from 229 billion in 2007 to 12 billion in 2008, and just 3 billion in 2009 (Figure 1). For all practical purposes, the CMBS market had ceased to function. Interventions designed to trigger new CMBS activity during this period, inc