First, Let’s Kill All the Servicers

As the housing/mortgage crisis has progressed, homeowner advocates and legislator have to get mortgage servicers to offer more loan modifications to struggling borrowers. Even though this housing recession has a far higher proportion of borrowers seriously underwater than past downturns, the logic of loss mitigation is still valid. It’s still better for the bank to keep the homeowner in place, even at a reduced payment, than foreclose (although in some communities where home valued have plummeted, the banks seem content for the moment not to take action against defaulting debtors).

Some observers have taken the view that it’s impractical for banks to negotiate on a case-by-case basis (gee, that’s how they used to make loans and do workouts before they decided to go for efficiency at the expense of quality). And that may be a valid objection: serivcers are factories. Even by some affordable housing experts report that they are unable to do one-offs.

Another impediment is that securitized transactions often limit the ability of the servicer to do loan modifications (although no one seems to have good estimates on how often that is operative).

Last year’s Treasury sponsored Hope Now Alliance was a cosmetic rescue program to address those issue, offering a “one-size-fits-few” template that was anticipated to offer servicers a legally defensible ground for making mods. However, few have happened.

Now legislators are considering another approach: require servicers to attempt loss mitigation before foreclosing. From Credit Slips:

n my prior post on mortgage servicing, I talked about the potential of mortgage servicers to be harmful barriers between homeowners and investors, both of whom may want to negotiate a loan modification. Recognizing such a problem raises the question of a solution. U.S. Representative Maxine Waters recently introduced legislation that would profoundly alter the duties of mortgage servicers. The bill, HR 5679, The Foreclosure Prevention and Sound Mortgage Servicing Act of 2008, would prohibit the initiation of a foreclosure if the mortagee or servicer has failed to engage in “reasonable loss mitigation activities.” The bill lays out exactly what counts as loss mitigation and offers up non-binding guidance on standards of affordability for loss mitigation. Servicers would have to report data on their loss mitigation activities, disaggregated by the type of mitigation activity (separately accounting for things like modifications, deeds in lieu of foreclosure, or repayment plans).

The bill also takes aim at the communication problems between servicers and homeowners. The bill requires services to provide a toll-free number that provides borrowers with direct access to a person with the information and authority to fully resolve issues related to loss mitigation and specifies that such a person must be physically located in the United States. Servicers are also required to forward borrower’s information to HUD-certified housing counselors whenever a borrower is 60 days or more overdue.

In the hearing last week on the bill (which you can watch as an archived webcast), Chairwoman Waters kept returning to a fundamental point–mortgage servicing is an unregulated industry. The witness testimony was essentially unanimous that mortgage servicing has a tremendous impact on American families and on the resolution of the current crisis. Of course, the debate was over whether this regulation was the right approach. The bill hasn’t gotten much publicity yet, but I encourage readers who are interested in the foreclosure crisis to take a look and post their feedback.

Mind you, I skimmed the text only quickly, but several thing stood out. This bill is clearly thought out, and is tough. Not only do the servicers have reporting obligations as discussed, but there are explicit requirements regarding communication with the borrower. For instance:

`(2) INCOME USED IN DETERMINING AFFORDABILITY- In making a determination of affordability for purposes of this subsection, a mortgagee or servicer shall use the income information furnished by the borrower at the time of loan origination, except that the borrower or mortgagor may elect to provide the mortgagee or servicer with current information and, if so provided, such current income information shall be used for purposes of determining affordability. The mortgagee or servicer shall advise the borrower or mortgagor of any right under this paragraph to provide current income information. If current income information is used, all sources of income shall be verified by tax returns, payroll receipts, bank records, or other third-party verification; the best and most appropriate documentation shall be used….

`(4) WRITTEN NOTIFICATION OF AFFORDABILITY CALCULATION- The mortgagee or servicer shall notify the borrower or mortgagor in writing of the results of the determination of affordability under this subsection and the income on which the determination was based. Such written notice shall be provided by mail not later than 7 business days after such action is taken or as part of the written notice required under subsection (c)(1), whichever is earlier.

The bill also permits servicers to recover “reasonable fees” (although they are subject to review) and applies to any “federally related mortgage loan that is secured by a lien on the principal residence of the borrower or mortgagor” that defaults after the bill is enacted.

My initial reaction is schadenfreude: the industry has brought this sort upon themselves. If it can’t figure out a way to do mods despite the mounting pressure to do so (and in cases where a mod might be possible, lose/lose to borrower and investor for failure to do so), it will be imposed on them, and as this bill delineates, in a way that offers them little wriggle room. My second is that it won’t be passed; there appears to be no Senate version of this bill. My third is that that’s a shame, the credible threat of the passage of a measure like this would light a fire under servicers (although that might only be to mobilize lobbying against it).

But to the more serious question: is this bill a good idea? Given the track record so far, it may be that servicers need something this confining to force them to act, and to give them cover with their investors to do so. Mods, however, typically favor certain tranches over others, so this might produce some sharp repricings.

Moreover, even if the team at the servicer has the best of intentions, it has to start with the mit plan from the income records at the time of the loan. With no docs, low docs, and generally lousy standards, that information is generally terrible. And while the debtor can provide current information, it isn’t clear they will be forthcoming. Servicers aren’t particularly well liked or trusted in borrower land.

But as much as I have little sympathy for companies that made a lot of money during the gravy days of the housing bubble and now plead poverty as an excuse for inaction, this bill would an industry under stress to the wall. It will have to incur the costs of notification and reporting for all defaulting borrowers, but will be able to collect fees only out of continuing cash flow from the borrower or a foreclosure sale. Most of these activities (setting up a call center, creating methods, forms, and supporting systems for borrower communication and regulatory reporting) have high fixed and low variable costs. It will be hard to determine the right fee levels up front (and you can’t readily go back if you got it wrong). And its quite possible the fees will look disproportionate, even if they are costed properly.

Plus there’s a big cash flow problem: the costs are incurred up front, the fees recovered over time. And servicers are already having big time cash flow problems. I have been told that servicers, which in most cases are part of large banks, are hemorrhaging cash. Notes from a conversation with informed parties:

The servicer has guaranteed to pay whatever interest is promised to the investors for at least 90 days after default. Some agreements also require them to pay principal during that period. Even after the 90 days, the servicer has to continue to pay real estate taxes and insurance. The servicer can use any late payment or other penalties from the borrower to offset these costs.

The idea was that if you were good at collecting and efficient at serciving, the overcollateralization would give you enough of a cushion. But they assumed 5% loss rates, not 20%.

And the cash flow drain is worse in prime or mixed pools than in subprime pools. The defaults relative to assumptions are far worse there.

As we noted, the banks will probably dodge this bullet. But a measure like this is an indicator of how sentiment about regulation is changing, Expect to see more tough-minded proposals, some of which will stick.

Print Friendly, PDF & Email


  1. Steve

    The servicer is a fiduciary of the note owner, no? If the servicers are, allegedly, failing to maximize NPV, then the note owners or collateral trustees have legal recourse against them. This sort of legislation is an attempt to break the mortgage contracts without explicitly doing so, and if we get laws like this it pretty much guarantees that consumer credit will become unavailable except to those who don’t need it.

  2. Yves Smith


    Agreed, but…..

    The servicer is the fiduciary of the owners of multiple tranches which have disparate economic interests. That’s why agreements that restrict mods require the approval of a majority (sometimes 2/3 or 3/4) of all classes to lift that restriction.

    Just like shareholders, investors are disenfranchised. Even if you suspect your servicer isn’t doing the best for you, if you sue, you incur the costs and everyone else is a free rider. It’s not worth it.

    Iif you are, say, a small business unless you have at least $300,000 at issue (unless you are talking small claims court stuff). You get to $50K in legal fast, and that’s without doing much in the way of discovery. If you are talking securities matters, where you have to hire big ticket law firms to be taken seriously, and in a case like this, you’d need to do a ton of forensics, I can see the legal costs getting very big very fast (those with real experience here encouraged to offer other estimates). It’s not worth it for a bondholder to incur the brain damage. Just as with stocks, it’s easier to sell.

    So you do have a real possible inefficiency here: servicers who can’t be policed.

    I have been told repeatedly that servicers are simply not able to do mods on an individual basis. So while I’ll agree that this proposal looks pretty extreme, it may well be possible that this will deliver better (or no worse in aggregate) results for the investors.

    We now have no way of knowing how much servicers are leaving on the table by not doing mods (I’m overstating this a tad; I recall Tanta and some press reports indicate that some lenders are at least offering repayment plans, which allows borrowers to catch up if they’ve missed a payment or two but that is not the same as a mod), But loss severities are 40% or more, from what I can tell, and that’s with a lot of delays in foreclosure processing, both due to capacity issues and banks not wanting the REO. So if you had foreclosures moving forward and property being put on the market, the real market clearing rate would be lower. 50% losses? 60% losses? With those kind of losses, some mods ought to make sense, but the servicer factory is set up to foreclose, not mod.

  3. Steve


    Thanks. You make a number of good points about the cost of resolving agency problems and about understaffing at servicers. Still, I’d add two things. First, if MBS investors actually believed that loan mods are a slam-dunk win-win and that servicers are hurting returns, Bill Gross wouldn’t shut up about it. What I hear coming from Pimco instead is cold horror at this sort of interference in contracts. Second, Joe Mason points to a Moody’s study that shows a best-case success rate of 52% on mod’d loans and an average success rate of 31-40%. Half or more of mod’d loans wind up back in default in under 24 months, which increases loss severities vs timely seizure of collateral. And those numbers don’t include the ridiculous credits issued in 2006-2007.

    I’d almost rather see assistance to debtors on a voucher basis than to risk the permanent closure of the mortgage market. But that would have too high a political price for Congress.

  4. eh

    I’d almost rather see assistance to debtors on a voucher basis than to risk the permanent closure of the mortgage market.

    Oh man.

    The thing that bugs me about “loss mitigation”, or whatever euphemism you want to use for it, is that it (nearly) always seems to involve measures that I see as benefiting the borrower, who IMO is often just a co-conspirator in what was self-evidently mortgage fraud on a rather large scale. And I don’t see anything all that “political” about the outrage that (I hope) would greet any suggestion to do such a thing.

  5. Richard Kline

    I like this bill rather a lot for a consumer protection armature for the reconstruction of the mortgage industry after the rubble is on the ground. I don’t see any way that these provisions can get traction on existing servicer and debtor problems.

    I do rather think that if the holders of mortgages cemented into MBSs don’t get into the act in doing mods, the decisions about _how_ the mods are done WILL be taken away from them. Now, I suspect one reason those MBS holders are dragging their feet on this descrying the sacred state bond of the mortgage contract is that they are hoping that _the MBSs in their entirety_ will be taken away from them at inflated values by The Powers That Be. This is one of the problems of our attempts to present asset price deflation, that the holders of these securities hope, pray, and believe that they will be substantially bailed out so they have every incentive _not_ to get involved in mods en mass, in all seriousness. If it ’twere made plain to the holders of these securities that they and they alone will take all losses from the underlying mortgages, they would have rather more incentive to GET ON WITH IT. But we don’t do things that way.

    ” . . . [L]aws like this it pretty much guarantees that consumer credit will become unavailable except to those who don’t need it.” Until the government floats publicly backed, nonprofit, low cost consumer credit unions to handle consumer finance at the bottom of the chain. Rather like Japan’s Post Office savings system no matter how configured. We don’t talk about this, but it’s something that we should.

  6. Anonymous

    Richard Kline wrote:

    “This is one of the problems of our attempts to present asset price deflation, that the holders of these securities hope, pray, and believe that they will be substantially bailed out so they have every incentive _not_ to get involved in mods en mass…”

    I wish that some knowlegeable person would explain how the 2005 bankruptcy laws are impacting this process–what the laws were before, what they were after and what the changes mean in the current environment.

  7. burnside

    I simply can’t credit the idea of servicers unable to perform loss mit on a case-by-case basis, particularly if supported only by assertions to that effect, however numerous they may be.

    When you read CR and (especially) Tanta posts on the subject, you may come to consider those assertions uninformed at best. Certainly, she provides perspective from about as close to the action as you’re likely to get.

    Then, the discussion seems always to devolve to whether or not investors in MBS should somehow ‘benefit’, when in fact they are exposed to two disagreeable outcomes, i.e., mitigation of terms or repo – neither of which is welcome news to an investor.

    There seem to be many, many questionable bases to this discussion and, however logical or valid it may appear, I suspect it is not correct in essence.

  8. bob

    In making a determination of affordability for purposes of this subsection, a mortgagee or servicer shall use the income information furnished by the borrower at the time of loan origination,

    Total bullshit. What are these people thinking? How can they be required to do this on a case-by-case basis and *not* go for full doc? It is insane.

  9. Pure Guesswork

    Doesn’t this situation open up an opportunity for a business that would do just what the servicers say they can’t do? Essentially to de-securitize these mortgages?

  10. Yves Smith


    I have heard that servicers are too automated, for want of a better word, from people who have every incentive to say otherwise, namely, people in affordable housing (and they have been very skeptical that Bush & Co. would do anything to intervene, so this isn’t a posture to force Federal intervention). The ones I have spoken to are very knowledgeable about MBS contracts and servicer procedures, so I take what they say seriously.

    I can’t recall in what context this came up (maybe an early version of the Hope Now discussion, when it appeared it might be more sweeping than it turned out to be) but Tanta said it would take six months of systems work to put it into effect.

    Maybe you haven’t dealt with credit card operations or other financial services processing businesses, but they ARE rigid, process-driven. If they do something different, they have to invent (well, get approved, nothing is simple) near-manual procedures. In a high volume environment, that is prone to error.

    And they lack the skills. How does a servicer determine who might be able to pay and who not? They don’t have a credit function. That was on the origination side (to the extent any assessment was done at all, now we know it wasn’t). There is scantly little information in these banks that would allow them to ascertain who might make it, once they have gone underwater, and who not. That may be why the record per Steve above is so lousy. My understanding is mortgage counselors have a better success record, but they do a lot in person and some of it is subjective assessment based on experience, just like in old-fashioned banking. That also doesn’t fit in a highly procedural environment.

  11. Yves Smith

    Anon of 9:21 AM,

    The short answer is that if you are not eligible for Chapter 7 (in most states, that means your income is above median), and the mortgage is non-recourse (true with firsts in most states, not true with refis) it is easier to walk from your house than your credit card debt.


    Did you read further? The customer can (should) correct it. If he inflated his income, the mod will come up with a payment schedule he can’t afford. or more likely, that no mod is warranted (ie, according to what he said, he ought to be able to pay). The borrower has an incentive to cooperate if he wants a mod.

  12. burnside

    Yves, my take on this was and is colored by Wednesday’s (4/23) discussion of the Foreclosure Prevention Working Group report at Calculated Risk, and particularly by two comments posted by MoM late in the discussion. It’s in-some-depth Tanta, so takes a few minutes to read.

    It made clear to me that applicants for mitigation and their counselors may not receive much more than the information contained in a notice of adverse action if their efforts to request mods fail. They’re entitled to their opinions, surely, and I can hardly blame them for finding the experience difficult, formal, impersonal.

    They may actually be getting fair treatment (or foul) at the hands of the servicer, but I’m not sure how they would know in either case.

  13. Yves Smith


    I will read the report, and I may be making a mistake by giving you my take first and reading the report second, but here goes (I sent much of this to reader Steve who was also kind enough to send along a paper by Joseph Mason).

    I’ll see if Mason gets at this, but my issue is that the failure of mods as currently practiced is not necessarily an indictment of mods in general. My suspicions go up when the prevailing practice in every past real estate downturn is suddenly declared no longer to work.

    Yes, in the current environment, probably only a fairly small percentage of defaulting borrowers might be salvageable. But my guess would be that that percentage is actually something like 20%. It might be even as high as 30%.

    Why? The loss severity on the real estate. The fact that servicers are not foreclosing says they know the losses on a sale would be pretty bad, or the properties would be close to unsaleable. So they are getting NO cash flow,, zip, nada. And if the borrower/occupant is not paying the taxes, it becomes a lien on the property. No savings to the investors of expenses by keeping the deadbeat in place (save property maintenance), mere deferral.

    With that fact set, the investors ought to be wiling to take a major writedown on the loan balance (or effective by keeping the nominal loan balance but having the rate be the one that would apply to a much reduced balance). It the loss severity would be 60% or more, that says writing the loan down 40-50% in economic terms would still be a win, particularly since the bank would escape the tax and insurance liabilities.

    So why aren’t they doing that? My guess is because investors and far more important, the bank parent have to take bigger writedowns on MBS and REO.

    Other big impediment is the loan servicers (I have been told this repeatedly) have NO knowledge of the borrower’s financial situation due to lack of due diligence when loan was extended, plus lousy loan files (ie, what little may have been known was not documented properly). So the servicer would have to gather info on the borrower (who is unlikely to trust the servicer) when they don’t have the skills (they don’t have a credit function) and the personnel who can engage in high touch. I’ve been told by multiple sources they are all losing money now big time; the last thing they want to do is incur more costs (but note here their interest differs from that of the investor).

    That means the borrowers are right to be upset with an impersonal process. Unless the servicer demanded and got W2s, recent tax returns, or other proof of income (ie, all that stuff they failed to do initially), the servicer is using either stale, unreliable info on the borrower and/or FICO, which we now know is of limited value.

    So the current process, as far as I am concerned, is garbage in, garbage out. Bad/non-existent info on the borrower, plus unwillingness to offer as generous a mod as they could given the alternatives (at least in the markets with serious declines), leads to mods probably being extended to the wrong people AND servicers trying to squeeze blood from a turnip, when more generous terms might work.

    Consider: given the costs of doing a mod, the servicer can be argued to have an incentive to see them fail on a small scale to prove they aren’t worth doing. However, that view is a bit too suspicious (never attribute to malice what can be explained by incompetence). But the point is I see far more downside (in terms of costs) than upside (pulling out fees for longer) from doing mods. Therefore, I see no reason for them to try to do them at all, or if pressured, particularly well.

    And if the lender is worried about giving the store away, have the reduction in payments be for 3-5 years, with them going back to the old level unless the borrower opens his kimono and agrees to whatever rate is devised then based on disclosure. Not on a blanket basis; the rate and length of the reduction would be borrower/property specific. But that takes work and involves costs.

    This is just a very rough idea, but an old-fashioned bank could implement something like that, a systems-driven one would find it impossible.

    I’m not coming at this from a bleeding heart liberal perspective; I’m looking at this from a historical perspective. Mods worked when banks knew their borrowers. It seem that THAT is what is missing in this equation.

  14. Yves Smith

    Readers may be assuming that the headline means I am in favor of killing servicers. The Shakespeare reference is actually double edged. In our modern context, it’s assumed he endorsed anti-lawyer sentiment. But in the context of the play, the character is fomenting a plot to end freedoms. Thus lawyers were seen as defenders of the common man.

    I don’t have such a generous view of servicers; their existence is part of a business model that has proven to be badly flawed. We are going to have to limp along with them as best we can.

  15. Mike Dillon


    As I noted over at CR – servicers lack of response to mods couldn’t have anything to do with the fact that they (servicers), depending on the wording of the PSAs, profit more the longer a borrower is kept in default status could it?

    Again, depending on the PSA, the servicer may also get to keep any mod fees as additional servicing compensation. But once a servicer mods a loan, it’s no longer in default and therefore the servicer’s cash cow is cut off.

    I’m not at all above believing that servicers are refraining from modifying loans for as long as possible simply in order to suck a borrower as dry as possible before modifying the loan…. But that’s just me and I’m cynical as hell – probably due to fighting an illegal foreclosure initiated by Fairbanks Capital n/k/a Select Portfolio Servicing. Probably why I don’t hold a very high opinion of Hope Now either….

  16. Yves Smith

    I hope everyone will bear with me. Blogger has locked my blog. This is really an indictment of what passes for technology at Google; if you look at the characteristics of spam blogs, I don’t see how they could have singled Naked Capitalism out. And they ought to have screened it against my Google ad revenues or my Feedburner traffic, both of which they can readily access. Or better yet, contact me.

    Worse, I am speaking on a first time panel of econbloggers at the Milken Conference next week. This sort of thing never happens at a good time, but this is particularly badly timed.

    If you have any ideas, aside from getting off Blogger and raising hell in Mountain View, they’d be very much appreciated.

    Please keep checking back….

  17. burnside

    Yves, I’m not sure we’re both talking about the same thing here.

    Your friends are able to afford their homes and are having difficulties with servicers – I’m guessing that’s something along the lines of receiving an NOD while looking at one’s tidy stack of cancelled checks? Their pursuit of redress won’t place them in contact with loss mitigation staff, i.e., they’re not trying to negotiate a short sale, a reduction of principal or a reduced interest figure if their homes are ‘affordable’ – or are they?

    Some of the commenters here are equating servicers with loss mitigation. They’re not the same pair of shoes, as I feel certain you know.

  18. Yves Smith


    I think you have conflated some of the comments.

    My friends are very much involved with affordable housing. One buddy does leadership counseling (pro bono) with some of the leading mortgage counseling groups. He has also worked with banks that have actually made money doing community development. He has a LOT of contacts with people who have been trying to come up with remedies to the housing mess. He has put me on to some of them directly. And the ones I have spoken to aren’t as airy fairy as the press makes them out to be.

    So no, my comments are not based on negative experiences from friends who had a bad time with their servicer, this is from mortgage counselors and people at banks that do low-income mortgage lending (yes, there are not-for-profit lenders to the subprime types who have done it very successfully, and have default/loss rates like old-style prime loans because, lo and behold, they actually screened the borrowers and helped them put together a budget to see if they could really afford the house).

    So my sources are people on the front line, and I take what they say seriously. They were involved in the Hope Now discussions, BTW, and were pretty disappointed with how it was handled.

Comments are closed.