Fighting Uncertainty With Uncertainty
True to the populist environment that characterizes the clear air turbulence we all have to fly through for the next n months, mortgage "cram-downs" are a hot topic. Why not? On the surface they seem like an easy fix. Even from a Chicago School of Law and Economics perspective they have a certain charm, dare I even say, a certain charismatic appeal? Who would ever argue that between the creditor (large international financial institution) and the debtor (middle class- but increasingly upscale- individual) it is the debtor that holds the title of "least cost avoider"? And, of course, there is political capital to be accumulated by lunging at the banks and piercing a lung with the rapier of "financial justice." These combine nicely to create a bitter amalgam that, as a topical salve, is about as likely to cure the patient's syphilis as bleeding him. To borrow from a more brazen, amusing and fictional cynic: "I think I'm on the point of spotting the flaw in this plan."
Two key points are often omitted (either via malice or sloth depending on the intellectual integrity of the advocate) when this discussion, lava like, bubbles up from some subterranean magma pocket that is the political compulsion to "Do something! Anything! Now!":
- Mortgage debt for a primary residence has heretofore been off limits to the bankruptcy judge's red pen
- Mortgages are (collateral in the actual property notwithstanding) non-recourse loans
Careful reflection on these facts will highlight a third: for the cautious lender collateral is (or should be) paramount.
Collateral Is King
In a way, this is a good thing. Collateral, in theory, is easier to model than borrower credit worthiness, risk of prepayment, risk of default, or the risk of refinancing (which some would classify as risk of prepayment). The fact that the collateral itself is the "last recourse" to the lender dovetails nicely with its status as an easier asset to evaluate the "intrinsic value" of. You can, if you have the discipline, put your hands on it, inspect it, touch it. In short, when everything else turns to custard, the lender can take possession of a physical asset that is (relatively) easy to assign an intrinsic value to- if even the most basic effort is made to do so. Given the many complexities of mortgage securities, simplifying elements are a good thing™ when all the ballots are counted.
Between a Rock and a Hard Place
Unfortunately, all of these points lead to a deeper and more complex background which, while complex, is critical to understand. It is worth pointing out, and in fact I have little hesitation emphasizing that, if such details are beneath your notice (as opposed to familiar ground) you simply have no business discussing the topic. Period.
The two diametrically opposed facets of the obligation (bankruptcy "immunity" and non-recourse status) on which mortgage backed securities are based have found their own equilibrium over several decades and play a critical role in pricing both individual mortgages and the securitized products they underpin. Historically, tampering with this sort of equilibrium ends badly for all concerned. This pithy observation on my part vastly understates the real and practical problems in the mortgage market presently, as never before has so much wealth been tied so closely to a single equilibrium. This is a consequence of the structure of the market for mortgage securities, and, accordingly, this warrants some discussion.
Agency(ies) Problems
The more recent development is the creeping divergence of underwriters and the ultimate debt holders. Capital requirements (like Basel and Basel II) require compliant institutions to maintain fixed capital percentages in proportion to some asset metric, such as Risk Weighted Assets ("RWA").1 While an in-depth examination of Basel I and Basel II requirements is beyond the scope of this entry, it is worth noting that these limitations (or indeed, any prudent capital requirements) made it impossible for even the largest underwriters to maintain even small fractions of the market for mortgage securities on their balance sheets. Even with the introduction of nearly limitless balance sheets to absorb these securities (read: GSEs)2 securitization was inevitable given the size of the mortgage market. Once you realize this, and appreciate the paradox of separating underwriting responsibility from the ultimate security holder, you can delve deeper into the implications of a highly securitized market.
Uniformity Means Opacity
Anyone who has even bothered to crack the spine of Liar's Poker3 (and, fortunately or unfortunately, this likely means anyone who has worked in finance) will recognize that mortgages are far too irregular in isolation to securitize without a great deal of work. In addition to the obvious impediments to continued cash flow like default risk, mortgages have complications like pre-payment risk and refinance risks to add to their valuation-frustrating uncertainties. The most cursory inspection will lay bare the fact that no two mortgages are actually alike, and that even the "intrinsic value" of the collateral offered is subject to the whim and competence (or lack thereof) of what is often a single, less than objective and counter-incentivized individual: the assessor.
As the size of the market grows, the pure practicalities of it make careful monitoring difficult to impossible. The level of opacity between investor and collateral (and remember in mortgages this is the most stable, most predictable element) becomes high.
So, we have a non-recourse loan, dependent entirely on collateral only for recovery in the event of default, we have dramatically increased the level of opacity between the asset and the ultimate security holder, and we have made intrinsic valuation practically difficult or we have disincentivized its proper application.
The Fatal Blow
The Wall Street Journal quips thus:
Lenders have long fought against cram downs, which they fear judges will use aggressively to shrink loans for troubled homeowners. That could boost losses not just for lenders, but for investors in mortgage securities. Some bankers also contend that allowing cram downs will cause mortgage interest rates on future loans to rise because lenders would demand larger yields to compensate them for greater risks.
The Mortgage Bankers Association reiterated its opposition to cram-down legislation. "Any borrower who cannot stay in their home under [the Obama foreclosure-prevention plan] will certainly not be helped by cram down," said David Kittle, chairman of the trade group.4
I was interested enough in this last assertion to want to test it. Consider this:
Anyone modifying a mortgage contract ex post has the following levers to pull:
- Reduction of total loan principal
- Modification (reduction) of interest rates
- Freezing of variable rates to fixed rates
- Modification (extension) of the term
- Modification (reduction or "temporary" elimination) of mandatory principal amortization payments
Of these, the most benign to a lender (it may even profit her) is an extension of the loan term- in the absence of other modifications. Yes, this will change the cashflow dynamics, but the increased interest payments should more than make of the difference in most cases if the extension is significant enough. I suspect this is also the least likely modification we can expect to see.
The real claws would be in interest rate and (worse/better) principal changes. Interest rate changes, however, aren't likely to be of much use to retail real-estate investors. Consider:
A 30 year $350,0005 mortgage at 7.00% is going to run about $2300.00 a month.
If somehow this is more than the magic 31% of income... let's say... 40% of income, then what sort of interest rate adjustment would we have to make?
That metric makes homeowner income $5,750 per month, or $69,000 per year, gross. Assuming no material change in income (and this is a big assumption, compounded by low or no documentation issues) we need to bring this down to $1,782 per month. That translates to a 4.55% interest rate or a reduction of 245 basis points.
Ouch.
On the principal side, you'd have to bring the outstanding loan down to $268,000 to hit 31%. That's an $82,000 gift to the borrower or about a 23% haircut for the lender just on the principal (I haven't bothered to do the cash flow loss over the term of the loan or run the full actuarial tables but the total interest paid is the real issue here- and this multiplier is what makes it politically easy to slash principal amounts in a cramdown hearing. It looks to the unwashed like the lender is only taking an $82,000 bath. This would be bad enough, but it is a small fraction of the damage). I submit that principal modification will be the most tempting, and therefore the most used, cramdown tactic. This is folly.
This is, obviously, a crude model. I've assumed the loan is brand new with a 30 year term, instead of some years old with, say, 26 years to go. I've assumed the loan was underwritten with a fixed interest rate. I've picked a random interest rate for purposes of discussion. I have ignored what collateral coverage might be in place (the legislation limits assistance to loans that are 80-105% of the homeowners equity- no word on how that will be determined, or by whom it will be determined). It is simple to see the level of violence this does to the lender. It's significant.
Now consider.. when you remove even the thinnest semblance of certainty for lenders when it comes to loan principal, what the impact will be on a security that has heretofore depended on that one stable variable: the intrinsic value of collateral.
How anyone can understand these facts and still think that cramdowns are a good idea is totally beyond my ability to comprehend. But, then, I don't live in Washington, D.C.
Dear Legislative Idiot:
The mortgage market is fucked. Every day that we pretend it isn't is a lie that will compound negative interest. Plan accordingly.
- 1. Astute finem respice readers will find instructive these FDIC provided instructions (108KB .pdf file) for "Calculation of Risk Weighted Assets" from 1998.
- 2. Proposing legislation to reduce or effectively eliminate Fannie and Freddie's capital requirements became such a routine practice for, among others, Congressman Frank, that no year after 1996 seemed to lack such an effort.
- 3. Liar's Poker: Rising Through the Wreckage on Wall Street, Michael Lewis (1990).
- 4. Pressure to Rework Mortgages Will Ripple Through Industry, The Wall Street Journal, February 19, 2009.
- 5. I pick this figure to land nicely below the $417,000-$729,750 cramdown limits in the present proposals.
Entry Rating:
