(This post is about a new venture at Ocwen that will likely be immaterial.)
On the latest conference call, Bill Erbey commented that Ocwen will be pursuing a new RMBS arbitrage strategy:
I believe that Ocwen has substantial opportunities to leverage our strong servicing capabilities by exercising cleanup calls, call rights or investing in existing private label RMBS tranches that we service. Most of RMBS securities we service have cleanup call provisions that allows the servicer to call the deal at par, typically when it has been paid down to 10% of the original unpaid principal balance. Notwithstanding slower prepayments, we see a steady stream of deals maturing in the next several years.
The opportunity results from the arbitrage of the underlying loans in REO being worth more than the securities. In other words, the whole is worth less than the sum of the parts. A condition precedent to our investment is our belief that Ocwen’s servicing creates strong cash flows for the securities overall. We’re building out this program and expect to be in the market purchasing securities in the next few months.
Perhaps some enterprising hedge funds will try to front run Ocwen.
First off, a disclaimer: I don’t understand RMBS very well so please take what I say with a grain of salt. Here’s what I understand…
Over time, a RMBS will decline in size. When the RMBS becomes very small, the fixed overhead costs of the RMBS structure will become a burden on the mortgage servicer. Many RMBS have provisions that allow the servicer to terminate the trust by paying a certain amount of money to the owners of the various tranches of a RMBS. That amount of money depends on the terms of that particular RMBS.
There are arbitrage opportunities between the following three values:
- The value of the mortgage loans.
- What the servicer has to pay to exercise its cleanup call provision. This is defined by the contracts that govern the RMBS. Note that the terms of the legal contracts vary.
- The market price of the various MBS tranches.
#1 versus #2
The Servicer can purchase the non-REO Mortgage Loans for the unpaid principal balance. The UPB may be lower than the fair value of the loan for reasons such as:
- Changes in interest rates. I believe many RMBS were originated when interest rates were higher.
- Changes in loan quality. As the borrower pays off the loan, the loan becomes substantially less risky because the borrower has more equity in the home.
- When the loan was originated, it was worth more than the original UPB to begin with.
There are also forces that work in the opposite direction. Ocwen has purchased MSRs to RMBS with extremely high levels of delinquencies because Ocwen is good at handling delinquencies. Such RMBS likely had terrible underwriting for many of the underlying loans (e.g. stated income fraud, negative amortization mortgages, etc.). Over time, many of the bad loans will turn into foreclosures, REO, HAMP modifications (which take the loan out of the mortgage pool), or loan modifications. In a loan modification, the servicing contract stipulates how much of the principal can be reduced. I don’t know how that interaction plays out in practice.
The Servicer generally has to purchase REO properties at fair market value or some value defined in the servicing contract. From skimming the prospectus of a 2007 Wells Fargo RMBS prospectus, it seems to me that the contract does not clearly specify how fair market value should be defined. Part of the document suggests that the servicer would perform a “market value analysis” that is based on “a drive-by appraisal or broker’s price opinion conducted by an independent appraiser and/or a broker from a network of real estate brokers”. Not all RMBS contracts are the same. I skimmed through the prospectus for a different MBS which stated that two appraisals should be taken and the higher value used. My understanding of the legal issues is not very strong so I don’t understand if there would be an attractive opportunity for the servicer to profit for certain RMBS where the servicer can take advantage of flaws/features in the contract.
#1 versus #3
I will start by pointing out that RMBS are insanely complex. Here’s why:
- The underlying mortgages are complex. The prepayment features in a mortgage creates prepayment risk.
- Some RMBS have exotic loan types such as negative amortization loans, option ARMs, balloon payments, etc.
- To understand the quality and therefore the value of a mortgage, you have to understand the underwriting of the mortgage. Underwriting can be complex.
- Many RMBS are diversified geographically. Each state has its own mortgage/foreclosure laws, which means that the costs of a foreclosure varies from state to state.
- The modification options and government subsidized modification options add another layer of complexity.
- Homes are difficult to value/appraise with precision.
- Some RMBS are sliced into a ridiculous number of tranches (e.g. 20+).
- The servicer may engage in actions that benefit the junior tranches over the senior tranches, or vice versa. There are potential conflicts of interests. If the servicer obtains unrealistically high appraisal values for the REO and holds out for high prices, the junior tranches will receive payments for principal and interest that they would otherwise not receive.
- There are various potential conflicts of interest between the RMBS investors and the servicer. A well-designed legal contract would minimize those conflicts.
- The REMIC structure imposes certain tax-related limitations.
- Some RMBS have “yield maintenance” agreements with minor counterparty risk.
- Sometimes the servicer gets to keep any late fees, prepayment penalty fees, transfer fees, and other fees.
- RMBS contracts do not have standardized terms.
I believe that Ocwen likely has an edge in valuing RMBS tranches given that few people understand these complex securities. I think that one of the reasons why the US financial crisis occurred is because many institutional investors and investment banks had exposure to things that they did not understand. Personally, I am not comfortable in my understanding of RMBS. Ocwen should have a second advantage in valuing the (junior) RMBS tranches because it has access to the individual mortgage contracts.
Creating value out of REO
Normally, I believe that Ocwen is incentivized to liquidate REO as quickly as possible. It has to advance the cost of:
- Unpaid principal and interest.
- Attorney fees for foreclosure.
- Eviction. Sometimes ex-borrowers continue to live in their homes.
- Costs to clean up the property and perform repairs. In rare cases, borrowers will strip the home of appliances and steal the copper wiring and plumbing (!!). In other cases, borrowers will intentionally damage the home.
- Costs to preserve the property, e.g. winterization.
- etc. etc.
The servicer has to pay for the financing for these advances (which it will not recoup) so it is incentivized to advance as little money as possible. This is somewhat offset against the desire to sell the property as quickly as possible, so the servicer should put up a minimal amount of repair work if it will help the property sell faster.
If Ocwen were the owner of the REO property, it would behave differently. It would spend more money repairing and cleaning up REO to profitably increase resale value.
The big picture
The two main drivers of value creation in this strategy are:
- Arbitraging junior MBS tranches against what their value should be.
- Squeezing more value out of REO properties by removing the disincentives placed on the servicer under most RMBS servicing contracts.
*Disclosure: Long OCN.