Last week we saw the release of the details of the Basel Committee on Banking Supervision (BCBS) rule on Fundamental Review of the Trading Book (FRTB). Started in 2012, put forth in concept in 2014 and now proposed in detail last week, the rule, should it stand, makes it largely uneconomical to securitize all sorts of lending assets. As such, should this rule stand as written, it would most likely go into effect in January of 2019 and make it onerous for large banks to securitize a variety of transactions. The riskier the asset class and the longer duration of the asset, the more capital required. Thus, while auto, credit card and other short duration assets will be severely impacted, asset classes like private label residential mortgages and commercial mortgages will almost come to a halt. Here is a quick overview of the changes and why it matters to banks.
In some metro markets, loans originated for securitization make up almost 50% of commercial real estate production and a material portion of commercial lending. For longer term, higher quality credits, market share is closer to 75%. What used to be the domain of investment banks such as Goldman Sachs and Bear Sterns, have now all migrated to banks due to their lower warehousing cost, ability to leverage, contacts in the market and infrastructure. Right now banks that securitize can either use the “Standardized Approach” (SSFA) or an “Internal Ratings-based” (model) approach (IRB) to assess capital required for a securitization. However, if a bank is going to sell the securitized assets, it needs to use the Standardized Approach, which takes into account default risk, market risk and a residual risk (the latter serving as a catch-all category for all other risks such as liquidity, operations and others). Now of those components, the market risk factor just got extremely punitive.
It used to be that a bank could calculate the potential market risk by looking at the current market volatility and then looking at the securities’ sensitivity to changes in interest rates. This “value-at-risk” methodology, which largely focuses just on interest rate risk and optionality, now gets replaced with a methodology often referred to as “enhanced delta risk.” This concept means that in addition to movement in interest rates, securitized loan positions held on balance sheet to sell will have capital set aside that also takes into account not only interest rate and convexity risk, but also for potential changes in the shape of the yield curve, foreign exchange risk, liquidity and credit spreads.
While all that sounds logical on paper, once you take into account all these additional risks and apply the level of capital that the regulators want you to hold, large banks will now have to hold multiple times more capital than before, hurting their return on equity. To give you a flavor of the magnitude of change, if we did a $100mm securitization of commercial loans, we would currently have to hold around $10mm to $20mm in capital depending on the structure and the historic performance of the asset types. Under these new rules, that amount jumps to a most probable minimum of about $50mm with $95mm being very common. The real kicker is that if we structure some lower rated or non-rated credit tranches we could now have to reserve capital that is greater than the sum of the complete tranche. That means it is possible that for our $100mm example, we could have to hold 140% of the total amount. Why would a bank ever choose to do that, as it would decimate their return on capital? In other words, as the rule stands, conduit lending in certain areas, such as commercial mortgages stops.
This rule would apply to not only new issues marketed for sale but also secondary issues that banks buy to sell in the future. The rule change would effectively dry up liquidity so only banks that purchase these securitizations would do so for their own portfolio and not with the intention to trade, which is most of the market now. Further, while there are a variety of mitigations to the rule, such as increasing spread, getting guarantees and so forth, they all end up resulting in an unworkable market.
We are not sure if these rules are intentional or not, but if they are we would not expect them to be changed significantly, and the result is the end of the many securitization markets as we know them today. Even if not intentional, if the calculation methodology gets adjusted, it now seems likely that the capital requirements will still go up significantly. The end result is large banks will slow the production of longer data assets because now they will have to portfolio most of their origination. This will result in larger spreads and less conduit activity which means less competition for community banks. Keep an eye out for updates, as by the end of next year, if there is not some relief large banks will start curtailing their securitization efforts and most community banks (those that don’t sell into securitization conduits) will be the beneficiaries.
Submitted by Chris Nichols on November 02, 2015