September 12, 2022 | In this issue of The Institutional Risk Analyst, we feature a conversation with David Stevens, a two decade veteran of the residential mortgage market. David served as Assistant Secretary of Housing and Federal Housing Commissioner for the United States Department of Housing and Urban Development and later head of the Mortgage Bankers Association until 2018. We focus on the latest capital rule for mortgage banks from federal regulators. We also talk about the proposed acquisition of Black Knight (BKI) by Intercontinental Exchange (ICE).
The IRA: David, thanks for taking the time during a very hectic period in residential mortgages. Our friends at Ginnie Mae and the Federal Housing Finance Agency disrupted everyone’s Labor Day holiday for the second year in a row by dropping a new eligibility rule for MBS issuers in mid-August. The latest version, which we’ve dubbed the “McCargo Rule,” originates inside the risk function of Ginnie Mae. The rule is far more draconian than the Basel III treatment of residential loans and mortgage servicing rights (MSRs) for commercial banks. What is your take on this bizarre curveball from the Biden Administration?
Stevens: I have seen the numbers from several large nonbank lenders and what the new Ginnie Mae capital rule from Alanna McCargo means for their servicing books. One CEO told me last week that the one-year implementation schedule will put just about every Ginnie Mae servicer out of business. Oddly, the Ginnie Mae rule focuses on MSRs, which have never caused a single issuer to fail. Yet we don’t impose higher capital on non-QM loans or exotic residuals, which have caused firms to fail in the past.
The IRA: Under the McCargo Rule, according to the work done by the trades, several large IMBs fail and most of the correspondent channel for smaller government lenders disappears. How does this serve the mission of HUD to assist low income borrowers? In the definition of "net worth," Ginnie Mae's risk function proposes to subtract the excess MSR from capital? This is totally nonsensical. We worked on the early iterations of Basel. This imitation is an embarrassment. Does Ginnie Mae want us all to get out of government lending?
Stevens: The Ginnie Mae issuer rule is pretty draconian. A 20bp move in MSR valuations under even a modest stress test puts most IMBs out of compliance with the rule, which is an event of default under their warehouse and debt indentures.
The IRA: Several large investors called us last week to ask if they should not be selling Ginnie Mae servicing before the McCargo Rule is made final. Without a change in the Ginnie Mae issuer rule proposed by President McCargo, we think many clients will attempt to exit investments in Ginnie Mae MSRs. A sale assumes a buyer, however. Do the people at Ginnie Mae understand how thin the secondary market is for government MSRs? You're lucky to have two bidders on most bulk sales. This proposal will kill the government MSR market.
Stevens: Ginnie Mae used an outside consultant for the issuer capital proposal, so the RBC model does not make sense. If you subtract excess MSRs from net assets and get a negative number, that obviously is a problem. Your proposal in NMN earlier this week for Ginnie Mae to allow a 50% weight for excess MSRs is constructive but does not help several large issuers. President McCargo is not technical at all. I doubt she went through this model in detail, issuer by issuer. Frankly, she and HUD Secretary Marcia Fudge seems to have been sold a bill of goods by the career staff at Ginnie Mae.
The IRA: Several issuers proposed an interesting idea after the NMN column was posted: Keep the proposed rule, as is, but allow two ways to meet the requirement: (1) The current rule with 6% adjusted net worth, exclude the excess MSR; or (2) require higher net worth of 9% and include the “excess MSR" and term unsecured debt with > 1 year to maturity in net worth calculation. Bank holding companies are allowed to use term unsecured debt to fund Tier 1 capital investments in bank affiliates. The table below illustrates the idea:
Stevens: The implicit message from Ginnie Mae seems to be that they want net worth for government issuers closer to 10% than the 6% minimum in the rule, so let’s just raise the net worth number for the handful of big issuers that own most of the servicing and declare success. The FHFA proposal with the add-ons for pipeline hedging, for example, is the same thing. Let's put the net worth number for complex, large issuers where everyone is comfortable.
The IRA: Agreed. The all-in cost of capital for most non-banks is in mid-double digits, so 9 or 10 percent minimum net worth for large issuers is fine. Moving right along from the sublime to the ridiculous, you’ve had a lot to say about the prospective acquisition of BKI by ICE. We wrote back in May of this year about the investment prospects ("Does ICE + Black Knight = Shareholder Value?"), which seem a stretch given ICE’s lofty P/E multiple. BKI is a higher risk, lower P/E multiple busines than ICE. But the folks at ICE make so much money on the financial markets that taking over residential mortgages perhaps makes sense? What’s the deal?
Stevens: This is one of the most high risk business takeovers that nobody is talking about. I am amazed that more institutions are not realizing what ICE + BKI implies for the industry, including OptimalBlue loan trading platform and the Encompass point-of-sale, and how much influence ICE will have on the entire mortgage finance system. Many issuers have strong views about this transaction, but the market power of ICE and BKI intimidates most into silence. There is naturally a bit of paranoia, which our industry is known for, and also some outright fear of what this combination could mean in terms of access to new technology. Lenders are intensely concerned about what could happen to them in terms of cost or service levels if they speak out against this transaction.
The IRA: Well, BKI is the incumbent monopoly in the conventional servicing market for IMBs clearly. The legacy Fiserv (FISV) product still has a defensible position in banks, but some players like Flagstar (FBC) adopted BKI's MSP platform due to large correspondent and sub-servicing operations. We really won’t see the “new” Sagent product derived from Mr. Cooper (COOP)’s IP for servicing until next year, according to the channel talk. Only the fact of some larger Ginnie Mae issuers using Sagent prevents BKI from having a pretty much total monopoly in IMBs now. How do you resist the irrational desire of the ICE octopus to own residential mortgage servicing?
Stevens: When I was at MBA, I could beat up the GSEs and set some bright lines in terms of leaving the secondary market to the lenders. But with ICE and BKI, the MBA is conflicted because the head of ICE is on their board! This is one reason that Scott Olson of Community Home Lenders of America, who you featured earlier, has come out swinging. The acquisition of Black Knight by ICE could adversely impact many of his smaller members, banks and IMBs that could never afford the BKI product.
The IRA: Over time, the smaller issuers represented by CHLA will have their market choices and data controlled by the ICE/BKI combination. This is more than a little ironic and also problematic for ICE. The Consumer Financial Protection Bureau, HUD and the FHFA are moving in the direction of mandating new, higher levels of data retention for lenders and servicers. We hear that Sagent will likely be able to meet the existing FHFA data retention standard by March. We hear in DC that BKI is actively lobbying FHFA to push the deadline for even the modest fair lending data required because of huge technology issues they will have coming into compliance. If ICE buys the Black Knight legacy platform, they will essentially control the mortgage business, front and back, correct?
Stevens: For the smaller issuers that don’t have large servicing books, the ICE/EllieMae/BlackKnight monopoly won’t really impact them directly, but the indirect effect could be profound. MBA is conflicted as we mentioned and they would never stand publicly against an acquisition by a major member even if it adversely impacted smaller mortgage lenders. But perhaps more interesting is the silence from the policy community. The major political parties are focused on the midterm election. Fortunately, the Anne Schnare piece in Housing Wire got a lot of attention on Capitol Hill.
The IRA: Do you think that the Biden Administration will challenge the transaction? After all, one of the biggest public mortgage issuers, PennyMac Financial Services (PFSI), is suing BKI for antitrust violations. How does this transaction get approved by the Department of Justice with this civil litigation ongoing? That aside, the negative implications of BKI for the ICE stock may be the biggest negative in the equation. There is a reason why FISV ran away from mortgage servicing as a business. Likewise, Fidelity National (FIS) ran away from BKI, of note. FISV and FIS are much higher multiple businesses than BKI or the Fiserv mortgage servicing business that is now Sagent. What are we missing here?
Stevens: You are correct. The ICE data business is not a people intensive business. Mortgage servicing, on the other hand, is entirely people intensive. The amount of manual effort involved in keeping clients compliant and following the rules from various agencies is astounding, even today a decade after Dodd-Frank and the 2012 National Mortgage Settlement. There are so many ways that BKI and ICE can get hung for being the owner of that platform in a recession, when loss mitigation is front and center for all issuers.
The IRA: When we left Kroll Bond Ratings in 2017 and became a consultant to Ocwen Financial (OCN), the CFPB basically told us to junk the Real Servicing platform in favor of the “better” Black Knight MSP platform. A decade later, we now have CFPB making noises around lender bias and servicing errors that seems to target BKI and other service providers for a future role as a responsible party in an enforcement action. And imagine if CFPB and/or FHFA mandate retention of all data used in the loan underwriting process. Significantly, Sagent does not see the consumer data as their property, but ICE and BKI do. Is that the bright future that awaits ICE if they acquire BKI?
Stevens: It is no secret that Black Knight has a mixed reputation in the industry with customers, but so far the regulators have not targeted BKI directly for errors in the lending or servicing process. The danger I see for ICE is that they have a lot more cash and a far better reputation than BKI, making them a prime target for ambitious regulators in Washington. The operational and political difficulty of running a massive loan servicing platform is a far cry from the automated systems at the New York Stock Exchange, where the audience is primarily institutional investors. With residential loan servicing, ICE will be directly facing millions of consumers and sharing in the risk of the largest mortgage issuers and servicers. The entire mortgage industry will have a single point of failure with the ICE/BKI combination, a monopoly that will stifle innovation and competition even as it concentrates operational and regulatory risk in one firm.
The IRA: It is amazing that BKI actually asserts ownership over legally protected consumer data in servicing. Most large Wall Street advisory shops that value loans and MSRs specifically avoid holding or claiming ownership of consumer specific data because of the legal liability. Just makes us want to run out and buy the ICE stock. Thanks David.