WORD ON THE STREET: Large, first-lien servicers have significant ownership interests in second liens and often have no ownership interest in the corresponding first-lien mortgage loans that are made to the same borrower and secured by the same property. In such cases, the first liens are typically held in private-label securitizations, and the second lien and the servicing rights are owned by the same party, often a large bank.
The four largest banks (Bank of America, Wells Fargo, JPMorgan Chase and Citigroup) collectively service 54% of all one- to four-family servicing in the U.S. They own approximately 40% ($408 billion out of $949 billion) of second liens and home equity lines of credit outstanding. The securitized second-lien market is very small. Thus, when a first lien in a private-label securitization is on a property that also has a second lien, that second lien is very likely to be held in a bank portfolio, and if it is inside a bank portfolio, it is often in one of the big four banks.
This is a conflict, because the servicer has a financial incentive to service the first lien to the benefit of the second-lien holder. Many times, this incentive conflicts with the financial interest of the investor or borrower. We outline some of the consequences of this conflict:
Consequence: Short sales and deeds-in-lieu are less likely to be approved.
An example makes this more intuitive. Assume that a borrower has a $200,000 first lien and a $30,000 second lien ($230,000 lien total) on a home that suffered a valuation reduction down to only $160,000. The borrower is paying on his second lien but not on the first lien. The borrower receives a short sale offer at the market value of the property, and asks the servicer (a large financial institution) to consider it.
If the servicer accepts the offer, the second lien (held on the balance sheet of the financial institution) must be written off immediately. If the servicer is also the second-lien holder, it may be more inclined to reject the short sale offer. In this case, accepting the short sale offer was clearly in the best interests of both borrower and first-lien investor. Similarly, a servicer will be less likely to accept a deed-in-lieu of foreclosure. We believe that national servicing standards should explicitly address this issue.
Consequence: Loan modification efforts are suboptimal.
Loan modification programs have two issues: They do not address the borrower's total debt burden, and they do not address a borrower's negative-equity position. As a result, the redefault rate has been enormous. We believe that both of these shortcomings share, at their core, one common trait: conflicted servicers. We look at each in turn.
Failing to address the total debt burden
In a loan modification, only the mortgage debt is affected. That is, most modification programs, including the government's Home Affordable Modification Program (HAMP), look at the payments on a borrower's first mortgage plus taxes and insurance, and compare that to the borrower's income. This is called the front-end debt-to-income ratio (DTI), and an attempt is made to reduce the payments to a pre-set percentage of the borrower's income. Consider a bank that services a borrower's first lien, second lien, credit card and auto loan. The first lien is in a private-label securitization, and all other debts are on a bank's balance sheet.
The bank is obligated to modify only the mortgage debt, leaving the credit card and auto debt intact. Moreover, the second-lien mortgage debt is generally treated pari passu with the first lien. There are situations in which only the first lien is modified and the second lien is kept intact, making even less impact on the borrower's total debt burden.
Since there is no sense of an overall debt restructuring, the borrower is often left with a mortgage payment that is affordable but a total debt burden that is not. For example, the U.S. Treasury Department's HAMP report shows that the borrowers who received permanent modifications under HAMP had their front-end DTI reduced from 45.3% to an affordable 31%, while the median back-end DTI (or total debt burden as a percent of income) was reduced from 79.3% before the modification to a still unsustainable 62.5% afterward. The result: a high redefault rate on modifications. For a successful modification, a borrower's total debt burden needs to be completely restructured.
Failing to address negative equity
Consider the Second Lien Modification Program, the HAMP program that applies to second liens. Essentially, this program treats the first- and second-lien holders pari passu when the borrower's first lien is modified. If there is a rate reduction on the first lien, there is also a rate reduction on the second lien; if there is a principal write-down on the first lien, the second lien also receives a principal write-down.
This makes no sense, as the junior lien is, by definition, subordinate to the first lien, and as such should be written off before the first lien suffers any loss. And if a modification is done outside of HAMP (and there are more non-HAMP or proprietary modifications than there are HAMP modifications), the servicer is not compelled to address second liens at all.
The negative-equity position of many borrowers would be dramatically improved if the second lien were eliminated or reduced more in line with the seniority of the lien. Indeed, loan modification programs would be markedly more successful if principal reduction were used on the first mortgage and the second lien were eliminated completely. Our research has shown that a principal-reduction modification has the highest likelihood of successfully rehabilitating a borrower, and will ultimately result in the lowest redefault
rate.
Principal reductions are used in loan modifications less frequently than they should be, due to conflicted servicers. Even with the current pari passu treatment on first and second liens, we believe there are fewer principal-reduction modifications on loans owned by private investors than there would be if a related entity of the servicer did not own the second lien.
That is, we believe banks are reluctant to take a write-down on a second lien that is paying and current; as a result, they do a first-lien modification, which is less effective, to the detriment of the borrower/homeowner, as well as to the private investors who own the first-lien loan. In addition, we believe conflicted servicers are counseling borrowers to remain current on their second liens, thereby allowing them to postpone the write-down on the second lien and increasing the likelihood of a pari passu modification.
Principal reductions are also used less frequently due to distortions in the compensation structure. Servicing fees are based on the outstanding principal balance. Thus, when a principal reduction is done, the servicing fee is reduced, as it is based on a lower principal amount. Since it costs more to service delinquent loans than the servicer is receiving in fees, and this is exacerbated by the write-down, it adds to the reluctance to do the principal write-down. With servicers trying to minimize the write-off of second-lien holdings and maintain servicing fees, it is no surprise that we see distorted outcomes for borrowers and investors in loans that banks service for private investors.
We can see a marked difference in servicing behavior for first liens owned by banks and those where the first lien is not owned by a bank portfolio. According to the Office of the Comptroller of the Currency (OCC) and the Office of Thrift Supervision's (OTS)Mortgage Metrics report for the fourth quarter of 2010, banks did a principal reduction on 17.8% of their first-lien portfolio loans.
These were loans in which they own the first lien, generally own the second lien (if there is one), and modified the first lien to achieve the highest net present value. By contrast, those same financial institutions did a principal reduction on only 1.8% of loans owned by private investors and 0% of Fannie Mae, Freddie Mac and government-guaranteed loans. While there are major obstacles to principal reduction in the case of government-sponsored enterprise (GSE) loans or government-guaranteed loans, there are few obstacles to doing principal reduction on private investor loans. Only a few pooling and servicing agreements prohibit such behavior.
And the OCC/OTS Mortgage Metric Report numbers for the fourth quarter of 2010 were not a fluke; in the immediately preceding calendar quarter, banks did principal reductions on 25.1% of their own loans, but on only 0.2% of loans owner by private investors.
Solution: increase the use of principal reductions as a loan modification tool.
National servicing standards should require that servicers perform the modification with the highest net present value, which will usually be a principal reduction. Under HAMP, the servicer is required to test the borrower for a modification using both the original HAMP waterfall, as well as the Principal Reduction Alternative (PRA), which moves principal reduction to the top of the waterfall.
If the PRA has the highest net present value, servicers are not obligated to use it. Use of the PRA is voluntary, at the discretion of the servicer. HAMP should be amended to require the use of the PRA, if it has the highest net present value of the alternatives tested.
Consequence of pari passu treatment of first and second liens: higher first-lien borrowing costs
We believe a large error was made in opting to treat the first and second liens pari passu for modification purposes. The consequence of this is that first mortgages will become more expensive, as investors realize they are less well protected than their lien priority would indicate. It is very important to realize that under present law and practices, a second mortgage can be added after the fact, without the first-lien investor even knowing it. But the addition of a second lien significantly increases the probability of default on
the first mortgage.
However, as presently constructed, if a borrower gets into trouble, the first and second mortgages are treated similarly for modification purposes. Since that raises the risk for the first-lien investor, it should also increase the cost of debt for the first-lien borrower. (We haven't seen this reflected in pricing yet, as few mortgages have been originated for securitization; most mortgages issued since the pari passu decision were insured either by the GSEs or the U.S. government.)
Solutions to maintain lien priority
What can be done about conflicts of interest inherent in an entity servicing a pool of loans and owning the second lien (while the first lien is owned by an outside investor)? There are at least three alternative solutions for newly originated mortgages.
The first two require congressional consent, while the third would require actions by the bank regulatory authorities. These solutions to the re-ordering of lien priorities are beyond the scope of national servicing standards.
Alternative No. 1: This solution would contractually require first-lien investors to approve any second lien (or alternatively, approve any second lien with a combined loan-to-value ratio (CLTV) exceeding a pre-set level, such as 80%). If the first-lien holder does not approve it, yet the borrower still takes out a second lien, the first lien must be paid off immediately (the "due on sale" clause is invoked). This may sound harsh, but it really is not. Currently, if a borrower wants to refinance his first lien, the second lien must explicitly agree to re-subordinate his lien.
The infrastructure to arrange these transactions exists and works smoothly. Prohibition of excessive indebtedness is common in corporate finance. This is done through loan covenants that limit the amount of junior debt that can be issued without the consent of the senior-note holders.
This alternative may be required to restart the private mortgage markets and would require an amendment to the Garn-St. Germain Depository Institutions Act of 1982. That act prohibits the senior-lien holder from invoking the due-on-sale clause if the borrower opts to place a second lien on the property.
Alternative No. 2: Place an outright prohibition on second mortgages where the CTLV exceeds a designated level, such as 80%, at the time of origination of the second lien.
Alternative No. 3: Establish a rule that a lender cannot service both the first and second liens while owning only the second lien.
Laurie Goodman is a senior managing director at Amherst Securities Group, a broker/dealer specializing in the trading of residential mortgage-backed securities. This article was adapted from testimony Goodman delivered on May 12 before the Senate Committee on Banking, Housing and Urban Affairs' Subcommittee on Housing, Transportation and Community Development. Her full testimony can be accessed here.
