Comment Rescue – The Regulation Edition

Dave

Dave is a part-time blogger that writes about whatever suits him at the time.

Related Post Roulette

76 Responses

  1. greginak says:

    Jumping straight to ** instead of staring with * seems to lack clarity and is confusing. ( insert smiley face here)Report

  2. Kim says:

    Sounds like someone actually thought about the incentives before issuing the rules.
    I like the sound of this.
    (I’m also encouraged that the banks are asking “please don’t hurt us for a typo!”
    rather than “please see if we can let a little fraud happen” — my experiences
    with financial companies haven’t always been the greatest, I admit to some bias).Report

  3. nevermoor says:

    First reaction: you assert that “(a blank line where a name is supposed to go is not the kind of inaccuracy that is appropriate grounds for a buyback)”.

    Depends on where that is. It might not impact the likelihood the loan performs, but it could very well make foreclosing the loan more difficult. And, of course, a non-foreclose-able loan is a big problem.Report

    • Dave in reply to nevermoor says:

      I updated my post and added an additional clarification. In the event of a data exclusion, significant or otherwise, prior to the initiation of a loan buyback, a lender should have the right to cure the defect by producing the correct information. If a lender fails to produce the information and the error is significant, then the GSE can pursue the appropriate remedies. I don’t see any problems with that.Report

      • nevermoor in reply to Dave says:

        I saw that. I’m just making the point that loan evaluation is two things: (1) likelihood the borrower is capable of repaying; and (2) likelihood that the security is forecloseable/sufficient to cover exposure.

        Lots of technical crap can be irrelevant for (1) while causing real problems for (2).Report

      • Dave in reply to Dave says:

        @nevermoor

        I’m just making the point that loan evaluation is two things: (1) likelihood the borrower is capable of repaying; and (2) likelihood that the security is forecloseable/sufficient to cover exposure.

        Lots of technical crap can be irrelevant for (1) while causing real problems for (2).

        I don’t think that (2) is really an issue with the loan evaluation process since the lender’s right to foreclose as a remedy to a borrower default should be written crystal clear in the loan document itself. That should be standard legal language.

        To me (2) seems like a chain of title issue but title should transfer to the GSEs. If I recall, the problems with loans not being able to be foreclosed upon had to do with the then current loan servicer not being able to produce clean titles for properties that they wanted to foreclose upon.

        Plus, any loan defaults within the first 3 years become the bank’s problem because the GSEs can force them to buy them back. Most defaults seem to occur during this time (IIRC) so it’s more of a problem for the banks.

        I’ll speak to your third reaction later. I have to head home.Report

      • nevermoor in reply to Dave says:

        @dave,

        All due respect but you’re wrong here. Foreclosing a property, especially after the completely B.S. summary procedures were squashed, requires proof that the loan was issued legally, that all formalities were observed, and that all state/local rules were honored. That’s a big part of what loan diligence is, not because a borrower is more likely do default if those things aren’t in the file, but because if they do the collateral will be difficult to get.

        Also, I’m 99% sure you are incorrect that loan default is a justification for buyback. Loan default is often the reason a loan-holder examines a loan for issues, but a valid loan that defaults should be sold back. That’s the whole risk driving the enterprise.Report

      • Dave in reply to Dave says:

        Also, I’m 99% sure you are incorrect that loan default is a justification for buyback. Loan default is often the reason a loan-holder examines a loan for issues, but a valid loan that defaults should be sold back. That’s the whole risk driving the enterprise.

        I’m confused. Did you mean shouldn’t be sold back? If not, I don’t see how we’re in disagreement here. Anyway, the source of my comment came from here:

        http://www.bloomberg.com/news/2014-05-13/fannie-freddie-overseer-easing-loan-buybacks-mortgages.html

        Banks will be freed of liability for mortgages with three years of steady payments even if borrowers send their checks late twice during that time. They’ll also be off the hook for loans that pass underwriting spot-checks before the three years are up. Fannie Mae and Freddie Mac will begin notifying lenders in writing when they’re relieved of responsibility for each loan.

        Fannie Mae and Freddie Mac also will stop automatically issuing repurchase requests when private mortgage insurers rescind coverage on loans. Such insurance is required when borrowers make a down payment of less than 20 percent.

        Both missed payments and a lack of mortgage insurance when such insurance is required is technically, at least according to my non-lawyer understanding, a default condition and the GSEs were issuing buyback requests when these instances took place. Plus, the spot-check language tells me that the banks are still on the hook.Report

      • nevermoor in reply to Dave says:

        Fair citation. I’m not sure what that means (does it mean that problems that could trigger a buyback no longer do if there are three years of payments or that less than three years of payments itself triggers a buyback).

        I would read it as the latter, but that is in part because I understand default risk on good loans to be the key investment risk GSEs take off banks’ books. That passage could certainly go either way.Report

  4. nevermoor says:

    Second reaction: I’m not sure I see the problem with regulations (and it looks like you don’t either – “there’s no need to repeal bad regulation in this case (the regulations are fine IMO)”). Given that, it looks like this is a call for more discretion in their application (when there are errors in a file, you want the GSEs to evaluate its severity before requesting a buyback). Seems like the opposite of a move to “make sure the rules are crystal clear” whether or not you agree with its merits.Report

    • Dave in reply to nevermoor says:

      @nevermoor

      This is a good point as well. I agree that we’re both on the same page with the regulations (I find them both necessary and fair).

      I guess I am arguing for more discretion, but I really didn’t see it that way. The reason to have the reps and warranties, the loan buybacks and life of loan exclusions are to protect the GSEs from what happened to them a few years after they became the biggest buyer of subprime crap in the market (they bought a ton from Countrywide). As we all know, there was so much blatant fraud, deceit and misrepresentation at almost every stage of the origination process. Hopefully, this will protect against the downside risk the GSEs have, especially once the loans are over three years old.

      Seems like the opposite of a move to “make sure the rules are crystal clear” whether or not you agree with its merits.

      In your “third reaction”, you tipped your hat to your bias (an understandable one given your leanings) so perhaps you are looking at it from the perspective of the GSEs. Fair point. I’ve dealt with bureaucratic real estate investors and lenders. With them, it’s a “check the box” mentality. If they see something that falls within the requirements, they run with it. From the GSE perspective, a data inaccuracy is a data inaccuracy. A data omission is a data omission. We can’t have those. Let’s make the banks buy them back!!!

      From a risk management standpoint, this is a massive headache because while banks can probably manage their risk with respect to how many loans will default over the first three years or how many loans may come back to them due to fraudulent/shoddy underwriting (I’d think that these loans would default too), how does a bank that employs hundreds if not more loan officers that originate thousands upon thousands of loans requiring millions upon millions (if not more) of data inputs manage the risk of having loans put back to them on the basis of immaterial clerical errors?

      I don’t think it muddies the waters to create a rule requiring a lender to provide Fannie or Freddie with the correct information in the event of missing or inaccurate data.

      I don’t think it requires too much discretion to examine significance if the question that has to be answered is whether or not the loan would have been originated in light of new information. If it is discovered that a borrower’s income is 1/3 of what was reported in the initial loan documentation or if the numbers on a credit score are flipped around (i.e. was 750 and now 570), I think the answers there are straight forward, especially in light of the Fannie/Freddie guidelines.Report

      • nevermoor in reply to Dave says:

        I’m not commenting on the merits of your suggestion (in this reaction anyway), just that it’s a move away from bright-line. I do agree (in theory) that an opportunity to cure is reasonable, though my experience in this arena (as a lawyer for two different “teams” at two different times) is that it’s easy to agree to cure problems but hard to actually do it.

        As far as how to manage the risk, when you’re dealing with huge samples the answer is that you identify a reasonable buy-back percentage and, if on any given deal you over-shoot that percentage you drill down to see why.

        Also, my other nit is that you propose a “whether or not the loan would have been originated in light of new information” standard. Standard should be “whether or not the GSE would have bought the loan.” The difference is that banks have an incentive to originate anything a GSE will buy.

        Strikes me that the ideal flow is:
        1. GSE identifies problem, determines whether problem is potentially curable.
        2. If curable, give bank X days to cure
        3. If not curable, or X days has passed, bank repurchases loan
        4. If curable, and bank submits cure, reevaluate loan with new informationReport

      • Jim Heffman in reply to Dave says:

        In a way, it’s another version of the “motte and bailey” reasoning that has been thrown around the nerdfringe lately, and while it’s been pointed out that it’s incorrect in its terminology, it’s still useful here.

        Everyone agrees that there needs to be regulation. That’s the bailey. But when regulators start saying things like “the address is correct in one place but incorrect in another, that’s a buyback”, that’s the motte. When they say “this paperwork is not completely filled out and I don’t care how much confusion it causes, if any at all”, that’s the motte. And when someone says “you know, is this really a good use of regulatory authority to ensure proper market function”, they retreat to the bailey and say “well I thought we all agreed that regulation is necessary, didn’t we?”Report

      • zic in reply to Dave says:

        +1, @nevermoor

        Nice and concise process. I’d put a step between steps 3 and 4, a review process, triggered if the bank can’t resolve it issues the GSE’s satisfaction but the bank feels it has met that requirement; paid for via an insurance fund, similar to deposit insurance.Report

      • Dave in reply to Dave says:

        Strikes me that the ideal flow is:
        1. GSE identifies problem, determines whether problem is potentially curable.
        2. If curable, give bank X days to cure
        3. If not curable, or X days has passed, bank repurchases loan
        4. If curable, and bank submits cure, reevaluate loan with new information

        This sounds reasonable to me.Report

      • Kim in reply to Dave says:

        nevermoor,
        also note that the suggested regs appear to want a different procedure for “banks with excessive errors” (which seems like a good way to prevent systemic issues from surfaces)Report

  5. nevermoor says:

    Third reaction: when a government regulator says “these loans are invalid and we won’t buy them” while a bank says “c’mon! the loans are fine! all issues are minor!”, I have a reflexive bias against the bank.

    I also find the argument that “The banks aren’t forced to sell to the GSEs and the GSEs rely heavily on the private sector in order to help fulfill their respective missions” unpersuasive. Banks need the GSEs so they can issue loans with greatly reduced risk. GSEs are created to facilitate that, but it doesn’t make them desperate for more loans. In any event, if GSEs aren’t worth it for banks to deal with, how are they getting so many loans? (e.g. slide 10 here)Report

  6. zic says:

    A review process would be a good idea; I’d probably fund it with something like deposit insurance. The more errors your bank produces, the higher your premiums. I’d allow document correction if it doesn’t shift the borrowers obligations; but pay attention to the different between document correction territory and to it’s-time-for-the-bank-to-negotiate-a-new-loan-with-this-customer-because-the-bank-f’d-up territory merits some consideration, too.

    Banks have been notoriously bad at helping customers with loan modifications, they haven’t put much good-faith effort into making sure their customers can plan their financial lives, either. That is a two-way street; you wanna get some, you otta give some else your create uncertainty in people’s lives.

    I also doubt this idea will get much traction, I’m guessing most people still want to see banks punished for what they did to us, more than they want to eliminate regulation, and the opportunity for liberal rule-easing lost. Sigh.Report

    • Jim Heffman in reply to zic says:

      “Banks have been notoriously bad at helping customers with loan modifications, they haven’t put much good-faith effort into making sure their customers can plan their financial lives, either. ”

      What Dave is pointing out is that this applies to the banks themselves, as regards government regulations. If you can’t ever be sure that you won’t have to eat every loan you sold to the government because some bro at FMA decided that dates should be “year month day” rather than “month day year”, then why would you get involved in that business beyond the legally-mandated minimum requirement?Report

      • nevermoor in reply to Jim Heffman says:

        …because that’s an absurd exaggeration, and your bank makes huge amounts of cash by originating loans and then selling them to GSEs. Unless you think that GSE’s got $1.7T of securitizable loans last year by trickeration. If banks wanted out of the business, they’d get out. What they want is to stay in on better terms. The question here is whether better terms are reasonable.

        End of the day, loan issuance is an inherently messy and complex process which means that files will be imperfect. The question is whose burden that should be, the bank that wrote the file or the GSE that bought the file based on a rep&warranty that the file was clean.Report

      • Patrick in reply to Jim Heffman says:

        It’s not quite that cut and dried.

        The guy or gal that is trying to get the loan carries a lot of the burden here, as well.Report

      • nevermoor in reply to Jim Heffman says:

        @patrick

        Sure. Of course they do. But as between the originating (or, at least, selling) bank making reps and the GSE buying based on reps, this discussion is about who is on the hook if/when the reps aren’t true.Report

      • Jim Heffman in reply to Jim Heffman says:

        “End of the day, loan issuance is an inherently messy and complex process which means that files will be imperfect. The question is whose burden that should be, the bank that wrote the file or the GSE that bought the file based on a rep&warranty that the file was clean.”

        I love how you just ignore the entire post, the thrust of which is that “clean” is increasingly subjective and God help you if today’s reviewer feels differently than yesterday’s.Report

      • nevermoor in reply to Jim Heffman says:

        @jim-heffman

        I understand that you reflexively support anything and everything bank-related. The accusation that I “ignore the entire post” despite the fact that I’ve engaged with nearly all of its content up and down the thread shows you aren’t worth engaging with.

        As a closing on this subthread, though, I’ll note that no one disputes there are problems with the rejected loans (and your suggestion that this is about date formats is invented from whole cloth). The dispute is over whether the problems matter to the loan. And the GSEs appear to be working on relaxing their definition of what matters.Report

      • Jim Heffman in reply to Jim Heffman says:

        ” I’ll note that no one disputes there are problems with the rejected loans (and your suggestion that this is about date formats is invented from whole cloth). ”

        Other than Dave, in the OP? Which I guess you kind of didn’t read all the way through before you started posting? But hey, I guess that’s what we should expect from someone who reflexively opposes anything and everything bank-related.Report

      • Dave in reply to Jim Heffman says:

        @jim-heffman

        Respectfully, I think @nevermoor has done more than his fair share of engaging my points. I’m not sure why you’re suggesting he’s not. I’m not disputing the fact that there are problems with the rejected loans. I’m suggesting that there’s a problem if loans are being rejected on the basis of errors and/or misrepresentations that can be “fixed” simply by lenders providing the correct information.

        Even if he has a reflexive bias against the banks by his own admission, he’s not reflexively biased against allowing lenders to cure errors in loan documents, as indicated by this comment here.Report

      • Jim Heffman in reply to Jim Heffman says:

        Fair enough, I guess.Report

      • Dave in reply to Jim Heffman says:

        @jim-heffman

        I try to be fair. It’s part of my charming personality.Report

  7. Michael Drew says:

    Not a big deal, but it seems to me that this isn’t a post that’s in the true spirit of the comment rescue. It’s a post of its own with a jumping-off point of a line in a comment.Report

    • James Hanley in reply to Michael Drew says:

      You are the pedant’s pedant for this blog.Report

      • Dave in reply to James Hanley says:

        Ok, kids…it’s all good!!! 😉Report

      • It’s not a pedantic point. It’s about what a comment rescue is, or what I thought I understood t to be, which is to raise someone’s writing in comments to a more visible place because it is worthy of consideration on its merits. And that’s ultimately about according respect, or in any case being clear about the relationship between this convention (“comment rescue”) and according respect. Which is ultimately just about how we treat each other.

        This post sort of steps all over whatever @zic had to say there (which IMO was not necessarily comment-rescue worthy, not that Zic doesn’t frequently write things that would be) on the way of getting to a (productive) agenda that the author clearly had completely separately from Zic’s point. I’d rather just see the point that Dave wants to make presented as a point worth making on its own. He could still use the line as the jumping-off point for that post. To me, to say it’s a comment rescue is to say that you think this is someone at one of her more formidable moments as a blog commenter (to me a charmingly humble commendation), saying something of depth and weight that’s worth pulling out so the community can consider and recognize it. I know it doesn’t always have to be like that, but in this instance it seemed to me that any sense that that’s the spirit of the thing had gone, and I wanted to register that I felt something might be getting lost. (Or not – I suppose I was also checking my own understanding of the convention against that of the community.)

        I care about this because the early encouragement I received to continue participating here often took the form of either comment rescues or suggestions that something I wrote should be a comment rescue (which then turned into suggestions that I submit guest posts, which then turned into admonitions that if I was going to write a comment of that length, would I please make it a guest post because, seriously, who has the time?).

        Others may completely disagree with me about what a comment rescue is. (I may have just completely perceived it wrongly per the years.) But I don’t consider it a pedantic point, personally.Report

      • …But that’s all just by way of explaining myself. I actually do understand seeing it as pedantic without that context, or even with it.

        …And don’t worry, Dave, I won’t go another round. I did want to explain myself, though. Mostly to you, in fact, as James was quite justified in being like, “WTH with this?”, as would you have been. So now you both have my explanation.Report

      • Dave in reply to James Hanley says:

        @michael-drew

        Thank you for explaining yourself. In some ways, even though we may not see completely eye to eye, your interpretation is not unreasonable. I think we’re good. and if you have any additional concerns, please feel free to reach out to me privately.

        None of this offended me.Report

      • Michael Drew in reply to James Hanley says:

        @dave

        Cool, thanks. One thing I wanted to be very clear about is that it’s not at all a big deal.Report

    • Bert The Turtle in reply to Michael Drew says:

      @michael-drew Are you proposing that Dave should buyback this post? It seems justified considering that he neglected to fill in his last name in the byline.Report

    • zic in reply to Michael Drew says:

      @michael-drew this stung a little. I also kinda thought the same thing, too. For instance, there are a lot of other areas where liberals have put fourth load lighteners (tax forms, anyone?), and there are a lot of strings on liberal programs that peeve people to no end, cost a lot of money to process, and exist to make sure we’re not helping the unworthy as we extend liberal largess. These points go unnoticed and unaddressed.Report

  8. clawback says:

    Or maybe we could assume both parties to these agreements read them, understand them and the legal environment in which they exist, and price them with the understanding that a buyback can be triggered by even seemingly trivial errors? Or does The Sanctity Of Contract only apply to eighteen-year-olds with a student loan app in front of them?Report

    • zic in reply to clawback says:

      Have you been to a closing lately? Did you understand all those papers?

      Most folks don’t, they just sign what they’re told they need to sign to make the deal happen, and they trust the banks lawyer, which they pay for, to look out for their interests.Report

      • James Hanley in reply to zic says:

        Aren’t the two parties he’s referring to the bank and the GSE, rather than the bank and the homebuyer?Report

      • zic in reply to zic says:

        Remedying errors in the paper will likely mean the home-buyer will have to play a role in this; and so has a legal interest that needs consideration. Certainly, the borrower should not bear the cost; they may have other rights (stemming from improper disclosure, I’d guess) that merit consideration. Not always, granted. But once in a while. It’s possible that some category of errors are egregious enough that they mortgages will have to be rewritten to allow clear title; so there are questions of interest rate changes, etc., as well. If it’s a new mortgage, I’d suggest the borrower get’s the cheaper interest rate based on the date or original issue and the date of re-issue. I was thinking of stuff like that.Report

      • nevermoor in reply to zic says:

        Forecloseability is not limited only to issues of clear title. And the remedies aren’t that simple.

        Imagine, for example, a state law limiting the closing costs on a loan to x% of that loan. You issue a loan for x%+$1. There’s no title question. But the loan is, as it stands, illegal and predatory. And good luck foreclosing on that loan.

        Now, there are often cure periods built into the law, or approved by courts, but the remedy isn’t “here’s your dollar back, now we’re suing to foreclose”Report

      • Dave in reply to zic says:

        @zic

        I refinanced my home in July. I understood every word of it. The lawyer that was present at my closing was my own, not the bank’s.Report

      • Mike Schilling in reply to zic says:

        And when I buy a computer, I know more about it than the person selling it to me does. That makes me an outlier.Report

      • Dave in reply to zic says:

        @nevermoor

        But the loan is, as it stands, illegal and predatory. And good luck foreclosing on that loan.

        Given that it’s an unacceptable loan product, one of the categories that qualifies as a life of loan exception, I don’t see this being a problem for anyone but the originating bank. The GSEs should have every right to send this one back.Report

      • Kim in reply to zic says:

        zic,
        no, they don’t pay for the bank’s lawyer. In fact, most homebuyers don’t pay for a lawyer at all, which is an idiotic move. The bank’s lawyer is not out for your interests, neither is the lawyer retained by either of the agents, both of whom have a massive conflict of interest.Report

      • nevermoor in reply to zic says:

        @dave

        We agree on that. The only point I’m making there is that foreclosability is a complex thing to assess, and not merely a question of clear title.Report

      • nevermoor in reply to zic says:

        @kim

        It just varies a lot state-by-state. I have an uncle in MA who is a lawyer doing that work, but here in CA we don’t use lawyers at all at closing. And, end of the day, the terms are non-negotiable at that point anyway, so I’m not clear what value a lawyer would add.

        In any event, the larger point is that it’s a state-by-state thing.Report

      • Kim in reply to zic says:

        nevermoor,
        Most states don’t require a lawyer.
        you should hire a lawyer first, because the agents have a conflict of interest, and are often willing to collude to get you to sign the contract.

        Yes, by LAW, you can’t negotiate at closing. But try telling an agent that, when they’re looking at their commission disappearing, and this maybe getting resolved a year later after arbitration, and maybe a court case. Particularly when your agent needed that money for Christmas (and keeping her own house).Report

      • Kim in reply to zic says:

        nevermoor,
        In short, the other guy is probably not a legal scholar. And people try to pull stupid shit at closing all the time.Report

      • nevermoor in reply to zic says:

        Fair enough. I certainly think more people should hire more lawyers for more things. And a lawyer would certainly prevent that type of thing from happening.

        I’ve just only ever done closings on purchases or refis (~5x in my life) with a travelling notary after getting the final termsheet in advance of closing, and that notary couldn’t care less about the terms of the documents. Plus, the docs are all preprinted so there wouldn’t be a way for her to try anything anyway. It sounds like my experience is different from yours (and perhaps atypical). I agree that the situation you describe calls for a lawyer.Report

      • Kim in reply to zic says:

        nevermoor,
        yeah, nothing like a lawyer to explain to everyone in the room what the consequences are for trying to renegotiate the price at closing.

        Then there are the REAL horror stories, like the house where nobody “cleaned up” after the dog….

        Or the Pgh story of a Midwesterner who had agreed earlier to pay for the “oil in the tank” not realizing that Pittsburgh doesn’t run on oil… (and thus he was buying, at overinflated prices, something he’d never need)…

        First time homebuyers get swindled all the time.

        (one of our requests was that the homeowners remove the buried effigy of the saint before leaving…)Report

  9. zic says:

    Dave, (and R Tod, too, please,) they can’t plan the risk because they lack data on which to plan; a history that X number of loans will be repurchased, no? It’s not long-term problem, first of all. Secondly, it’s completely dependent on the banks quality-control standards; at least the shoddy-paperwork portion. So while I understand that it makes risk-management planning difficult, it’s also an admission they don’t know how shoddy their paper work (and underwriting sandards) are. While I support some room in cleaning up shoddy rubber-stamping practices that took root in both the buying frenzy and foreclosing frenzy that followed, I don’t think the it’s an unknown that should stir a lot of pity. Yes, it creates uncertainty; but that amount of uncertainty is under the banks control. GSE shouldn’t be able to send back a mortgage they did right, so I would argue banks should have some method of showing they did it right or corrected problems via a 3rd party review.Report

    • Patrick in reply to zic says:

      Yes, it creates uncertainty; but that amount of uncertainty is under the banks control.

      That’s actually something of the problem right now. The banks feel like too much of the uncertainty is under their control. As a result, they’re overcompensating… the documentation process for what used to be fairly routine, stable, run-of-the-mill 30 year fixed loans is now stupidly ridiculous (I have a lot of relatives in the mortgage business).Report

      • Dave in reply to Patrick says:

        @patrick

        the documentation process for what used to be fairly routine, stable, run-of-the-mill 30 year fixed loans is now stupidly ridiculous (I have a lot of relatives in the mortgage business).

        My mom works for the lawyer that handled my closing and she can tell you the stories about the stunts banks pull at the last minute with the documentation, especially if their Fannie loans (given the requirements under the program).

        At the same time, they’ve attempted to simplify the process. My binder has these summary pages that I guess are now required under law to show all the economics of the deal in the most user friendly way possible. Whether or not people will get that I don’t know. My perspective is biased since I understand how mortgages work.Report

      • Kim in reply to Patrick says:

        Patrick,
        and that has nothing to do with loans being unprofitable to banks, so they don’t want to make them? (discl: this was true a year or two ago–I have not checked current rates).Report

      • Patrick in reply to Patrick says:

        I think if you think that banks are an efficient rational organization you probably haven’t worked for many large banks that have overlapping business units.

        Or ever, in the loan business. I know a *lot* of people who are in the area of the business where stuff gets done… the underwriters, the processors, the shippers, etc. I know considerable numbers of them that would gladly haul out the loan officers and the upper management and hang them from a tree and shoot them.

        There are a huge number of internal perverse incentives in the mortgage biz. If you don’t understand that, attempting to steer the mortgage biz via regulatory nudges is going to result in all sorts of bad outcomes.Report

  10. Mike Schilling says:

    It is completely wrong of the government to exploit an ambiguity to gain advantage over an organization which, if the situation were reversed, would never do anything like that.Report

  11. Mad Rocket Scientist says:

    Regarding the paperwork (especially since I just closed on a new home two weeks ago):

    It seems to me that a lot of the paperwork errors could be reduced or eliminated with the use of electronic forms. The mortgage I just got had a considerable amount of digital paperwork, with only the final closing involving a notary & signatures on paper, so all the forms will filled out electronically. If this is how it’s done, the forms themselves can be created with logic to ensure that data is entered on the correct line, etc., and that in the event that requested information is optional, checkboxes can be used to at least indicate that someone looked at the field & made a choice to not enter data.

    This is one area where government often seems reluctant to upgrade, especially when it comes to forms where completeness of information is critical (like 4478, the NICS form – a digital version would reduce errors and omissions by a large amount, but the BATFE refuses to offer one).Report

    • Lyle in reply to Mad Rocket Scientist says:

      Actually the forms are filled out on a computer, its just a matter of programing the .pdf properly. Here is a link to a pointer: https://forums.adobe.com/thread/693580?start=0&tstart=0 that shows this is doable.
      However it takes precious IT time and the It departments budget has just been cut again. One can even buy online copies of the forms to fill out. But as in the boom times folks are just plain to busy to do things right. Following the old line” there is never enough time to do it right but plenty of time to do it over”Report

    • Patrick in reply to Mad Rocket Scientist says:

      This is actually how some mortgage companies get their docs, even when they use paper. You fill out N forms, those forms go to a company that specializes in docs, they input the results of those forms, asking for corrections/clarifications if there are inconsistencies, they print out everything and ship back a packet of documents for people to sign.

      The real problem comes with discrepancies that aren’t the type that can be embedded in that sort of process.

      Here’s an example. I’m getting the specific details incorrect, but it will illustrate the point.

      Prior to 2000, you needed a set of information for income verification. A bunch of supporting documentation that showed whether or not a buyer could reasonably be expected to pay back his or her loan.

      After 2000, a lot of those rules were changed to veer to port, for a number of reasons.

      After 2007, a lot of those rules veered hard back to starboard, well past where they were in 2000.

      So let’s say you’re an author, and you get most of your income from book advances. Over the last 10 years, you reliably get about a million dollars a year in book advances because your last name is King. You want a loan.

      Prior to 2000, they’d look at your income average over those ten years and say, “Jesus, King wants a loan, rubber stamp that sucker and gimme my points, this guy is a mortal lock for paying back his loan.”

      After 2000, a whole bunch of guys that had income streams that were irregular like King’s suddenly get an okay to get a loan where… quite likely… their actual chance at maintaining that income stream was not great (because they were flippers, for one example).

      After 2007, the documentation requirements for showing that irregular income (basically, anything that doesn’t come from a salaried position where you’ve been sitting behind Corporation X’s desk for 5 years or more) has gotten so stringent it’s hard for anybody who has irregular income to get the loan. Because the documentation requirements are controlled by the underwriting department, and the underwriting department is getting a huge push from the head of the underwriting department to make sure that nothing they approve will ever come back from Fannie or Freddie as, “insufficiently documented, buy this loan back”, because then the head of the underwriting department loses their job.

      The underwriters are more risk averse now than they were in 1999. The loan officers are as risk averse as they have ever been (not very). Freddie and Fannie have Congress up their butt, so their idea of “sufficient documentation” is going to be driven largely by “are we producing enough of an appearance at diligence that all these folks crawling up our butt will go away”, which is a moving target.

      Everybody’s trying to hit moving targets.

      This doesn’t matter to anybody who is paying cash, and there’s enough incoming foreign funds that still the housing recovery is largely being carried by foreign investors paying cash, so the old “median income to home value” ratios for solid loans that applied from 1945 to 1999 are still useless for guessing what a house is worth in a given geographic area. So the real estate market looks good, even though it probably isn’t, still.

      This doesn’t matter to anybody who has a 800+ credit score and 5 years on the job for two breadwinners in a dual income household at salaried positions for Very Established Corporation.

      It matters a huge amount to anybody who doesn’t work in corporate America, because right now they can’t get a friggin’ loan.Report

      • Kim in reply to Patrick says:

        Can you drop me a cite that it’s foreign investors?
        I don’t see much of that around here, and my limited reading on Detroit suggests that it’s mainly wall street…

        Many people close to home speculate on housing, particularly when renting is so much more expensive than buying houses.Report

      • Mad Rocket Scientist in reply to Patrick says:

        I will admit that this time, the documentation I had to provide was voluminous. I was about ready to ask if they wanted to give me a colonoscopy, while they were at it.Report

      • Will Truman in reply to Patrick says:

        No joke. It was all one huge scramble. Since I am a SAHD I was able to do it, but it would have been next to difficult to meet the deadlines if we were both working. One thing after another, it was. and we still had a delayed closing because they couldn’t get the employment verification docs from Clancy’s jobs. I was making calls all across the state.

        The kicker was that the documents they were demanding demonstrated that, two weeks after closing, her employment contract ended. They didn’t care about that. They just cared that her employment was verified.

        There was also an issue of Verification of Rent, where they kept going back and forth between saying that they needed one year’s worth of rent check copies to two years.Report

      • Patrick in reply to Patrick says:

        Read Dr. Housing Bubble (latest relevant post: http://www.doctorhousingbubble.com/all-cash-buyers-investors-all-cash-real-estate-pulling-back-trend/)

        Investors are hungry for deals. Many were aiming to purchase distressed properties and this is where the bulk of the money was going. Big investors were buying up many of the homes in places like the Central Valley of California, Arizona, Nevada, and Florida. These markets were great for turning homes into rentals since the rental household trend was growing dramatically.

        This is changing, but the presence of so many cash offers meant that strangling the loan process didn’t kill the real estate market, which it otherwise (would) have.

        Detroit is a dead zone. So are many other real estate markets. The loan market is largely driven by the markets where people make money buying and selling real estate.Report

      • Mad Rocket Scientist in reply to Patrick says:

        @will-truman

        Our mortgage company did everything online. While the amount of documentation was frustrating, they at least made the process as painless as possible. We could scan & upload paper copies, and if we had an original digital copy, we could just upload that. So I’d get an email saying “We need to see your 401K statement.”, and I could download the statement from my 401K site & upload it to them & it was done, took just a few minutes.

        As for investors, the first two places we put offers on were both hit with competing offers that were putting half down & waiving inspections. Very few normal families can show up with $200+K for a down payment.Report

      • Mike Schilling in reply to Patrick says:

        The kicker was that the documents they were demanding demonstrated that, two weeks after closing, her employment contract ended. They didn’t care about that. They just cared that her employment was verified.

        This reminds me of a story Donald Westlake used to tell [1]. He got so many speeding tickets that New York state suspend his license for three years. When he finally got it back and went to buy car insurance, he got a good drivers’ discount, for having no tickets in the past three years.

        1. The mystery writer, and if the name’s unfamiliar, you should remedy that immediately.Report

      • Lyle in reply to Patrick says:

        A question, does anyone know how they dealt with the situation in the 1970s before automated underwriting and desktop computers (yes there were computers in the back room but not on the front lines). In particular what were the income validation requirements back then. I do recall having to provide tax returns and w-2s in 1978 (the one time I got a mortgage, not counting a first home improvement that was less than 10% of the house price).Report

    • What Patrick said. There is an industry surrounding the automation of paperwork.Report