This morning on the way to work I was listening to Psychic Andrea on my favorite radio station and she was talking to a woman who served as strawman on a couple of condo purchases and now she is stuck with them because the actual purchaser changed his mind. Psychic Andrea saw a visit to an attorney's office for the lady but you don't really have to be psychic to predict that.
Jesus all jokes aside, that really is evil. I'm guessing that the number of people who are in foreclosure and who seek out the freddie mac website is going to be pitifully small sadly.
I'm guessing that the number of people who are in foreclosure and who seek out the freddie mac website is going to be pitifully small sadly.
That's exactly why Freddie gives this to anyone who will take it and show it: churches, community groups, schools, the people next to them on the train.
I'm just doing my part in the "viral marketing" department here.
Tanta,
People seem to generally understand that Excel doesn't write your story problems for you, but not so with FICO.
Who do you consider most culpable in accepting FICO as a predictive tool? Were the ratings agencies the dogs who were supposed to bark? Did the original marketing present FICO as a proven and powerful tool?
To bad they didn't do a PSA about 5 years ago warning people about the first batch of fast talking people in suits.
Start something like this: Have you submitted a request for a loan approval. These requests are on file at the credit agencies and mortgage brokers make copies. They then call on you with a large stack of papers and talk about low rates and interest deductions. These brokers get large fees and you are left with an unaffordable loan.
Bankruptcies jump 40 percent in 2007
The American Bankruptcy Institute blames the mortgage crisis for heavy debt load, warns that this year could see more bankruptcies.
"NEW YORK (CNNMoney.com) -- The number of Americans filing for consumer bankruptcy increased by nearly 40 percent in 2007, according to the American Bankruptcy Institute.
In a report released Thursday, the ABI said that the number of overall consumer bankruptcy filings reached 801,840 last year, compared to 573,203 in 2006.
"The roughly 40 percent spike in consumer bankruptcies during 2007 presages [an] even higher [number of] filings this year, as the heavy consumer debt load is made worse by the home mortgage crisis," predicted ABI Executive Director Samuel J. Gerdano.
However, the report also showed that the number of bankruptcy filings declined 7.5 percent in December from November. And Chapter 13 filings - those available to individuals with regular income whose debts do not exceed specific amounts - also showed a decline from November to December."
Who do you consider most culpable in accepting FICO as a predictive tool? Were the ratings agencies the dogs who were supposed to bark? Did the original marketing present FICO as a proven and powerful tool?
That's rather a long story. Also, it's important to distinguish between using FICOs to determine the borrower's eligibility for certain loan terms, and using FICOs to establish basic creditworthiness instead of a full-file review. Those two aren't the same thing.
Originally, FICOs got into the mortgage business via the GSEs. They asked seller/servicers to collect and report them on every loan sold, but they weren't used for any particular purpose by the originator. The idea here was that the GSEs have huge databases of loans, so they could do some "calibrating" of standard loan approvals and FICO scores, to see if they were in fact predictive for mortgage credit (not just consumer credit, which they were designed for).
After several years of data-gathering, the GSEs wrote guidelines to use FICOs to establish due diligence levels (again, not to approve loans with). This meant that if you had FICOs in the lowest bucket, you had to do the most loan-level diligence (verify everything, look at every little detail of the credit report, etc.). In the middle bucket, you could do "normal" diligence. In the top bucket you could do "expedited" review. The idea there was that a high enough FICO creates a presumption of creditworthiness, and so you just looked for anything that might change the presumption. (That's different from the old paranoid way of doing loans, where everybody's presumed to be a deadbeat until you find some reason to qualify them.)
That approach didn't change until the GSEs started expanding their offerings, especially deeper into "flexible" products. Suddenly there were new "options" for borrowers, and so FICOs became a way of measuring eligibility. For instance, the GSEs upped their maximum LTVs for cash-out refis (it had been 75% forever), but only for borrowers who met "enhanced" eligibility criteria. FICO was an easy way to set a separate, overlaid set of criteria.
That's the use of FICOs that the Alt-A and second lien and subprime people picked up. The big problem for them was that there wasn't a bit of a gap between their "standard" loans and their "enhanced" loans, as there was for the GSEs. There was a great yawning chasm. So inevitably FICO started to carry more weight than it was designed to handle.
The rating agencies contributed to the problem because of their preference for quantifiable numbers on data tapes instead of file-level due diligence.
During this whole period, the pricing people also latched onto FICO as a way of making risk-based pricing distinctions. It's hard to do loan-level price add-ons with a computer by using the results of an old-fashioned UW review. But FICO? You can automate the whole process of putting a rate and points and fees on a loan.
Finally, of course, FICOs stopped "calibrating" with LTV (borrower equity) as a default predictor because we went through a boom in which everybody always had equity--even if you didn't have any when you bought, you had some in six months. You can't take performance of loans in a historical period like that and "calibrate" FICOs meaningfully any more than you can calibrate zodiac signs or SAT scores.
But there have, all along, been lenders who doggedly kept using FICOs only in the work-bucketing or eligibility/pricing distinction fashion, not as a core underwriting determinant. You can tell who they are because they're not in BK yet.
No suggestion, apparently, that the lenders should get out in front here and provide some service for their clients. Looks like the real news is that there's a moral hazard in disintermediation.
Did Freddie 'educate' the people two years back about mortgage scams in the same manner? Do you wonder why not?
Most foreclosed houses do not have much equity left after it goes to the bank and gets sold in auction. The money taken away by the foreclosure con-artist hits the pocket book of lender, not the ex-homeowner....now you see why they are concerned.
"This new YSP disclosure stinks. This may put a damper on bumping the rate up an 1/8 or 1/4 right before closing to get a little extra YSP. They keep coming up with new ways to limit our incomes."
So you get to closing and the broker hits you with "Oh you're rate went up" and you're in a real bind and have to close anyways. What they really are doing is pocketing the money. Just wow.
Did Freddie 'educate' the people two years back about mortgage scams in the same manner? Do you wonder why not?
To be fair, Fannie and Freddie have put money into consumer anti-fraud educational efforts for years. It all just got drowned out by the unending chorus of cheerleaders.
A few years ago, when prices were going up 50% a year, Fannie and Freddie couldn`t come up with some sort of alert. I guess it was very hard to figure that one out. But somehow they figured out how to get involved in "derivatives" and other weird things. (I said a long time ago on this blog that I expect residential RE to tank at least 60%(Florida) in the next 2-3 years.)
I'm am wondering if FICO could lose some of its predictive value since it is so well known.
This is 50% of what Suze Orman talks about. It sounds like you could be under a decent amount of financial pressure and fico optimize, as it were. Or you could be in roughly the same position and trash your score simply based on following her strategies.
I work with a complex ontologically based scoring matrix. Nowhere near as well funded as FICO, but in my experience they're great as a tool to inform people who already know wtf they're doing. Once you start hiring people who just live and die by the output file you're about 2-5 years from a shit blizzard (technical term) of problems.
If the underlying reality of the situation being measured changes the system doesn't know that. "Insignificant" variations in the nature of what your data feed reports can produce results that are mind blowingly wrong.
I'm sure the FICO people know this and hire alot of people smarter than me and give them more funding than I can dream of to account for all of this. But in the end there is always going to be the fact that every system is built on a paradigm of historical experience and if reality changes away from that too quickly your system will fail.
Also our sales staff regularly oversells what we actually can do to the point you'd think our existence would negate the necessity for several federal agencies to exist. That's what they're paid for I guess. But I often talk to large customers who think that our engine decides what reality is at any one moment rather than doing a damned good job at quantifying it.
But we do fuck up, more than we ever admit.
So yeah, the fact that FICO is apparently the financial version of St. Peter's ledger rankles me more than a bit.
"you are not going to be in a better situation because some other fast-talking guy in a suit . . . "
Or because some other fast-talking guy in a suit with a beard, a PhD, and the title of "Chairman" asks for some Congressional intervention to get us out of this mess.
It sounds like you could be under a decent amount of financial pressure and fico optimize, as it were.
Although one can do short-term gaming of FICOs, I actually think it's less possible or practical than many people do. Like those many people who for only $99.95 will sell you the secret of manipulating your FICO (hint: take the Ginzu knives instead).
The problem, for me, is that we pretend that there's a meaningful distinction in default probability in very small FICO increments. Frankly, the "over 620 OK" "under 620 not OK" is probably quite meaningful. Everything else is, mortgage-wise, closer to talking to yourself.
We pretend that it makes some sort of difference to have a 95% loan with a 720 FICO on one hand and a 95% loan with a 680 FICO on the other. As far as I'm concerned, there's no meaningful difference there. The 680 borrower might be very marginally more likely to get into too much debt and have trouble making payments, but both borrowers can equally easily lose their jobs and both can damned well walk away from upside down loans. As long as they have equity and employment, both of those borrowers will repay you. You might actually make more money, end of the day, on the 680 because while that borrower will end up repaying you, he will also pay you a couple of late fees in there.
Now, a borrower with a 720 today who had a 720 six months ago, on one hand, and a borrower with a 680 today who had a 720 six months ago, might be a different kettle of fish. But we don't look at trends, just today's snapshot. You have to root through the tradelines on the credit report like an old dinosaur underwriter to find trend.
There just isn't any FICO score at which down payment and ability to service debt doesn't matter. It's this last bit of stupidity that is getting the comeuppance, but still there were a lot of distinctions made during the boom that just didn't matter. That's why people are tempted to game their FICOs: they want a 720 instead of a 700 because you get the next price break at 720.
If we made the price-increments bigger we wouldn't have that problem, because you can't usually get enough FICO movement with cosmetic changes in your debts/limits etc. to move your score that much. In other words, if we went back to basing pricing on LTV, DTI etc. and just made the pricing difference under/over 620, the gaming would stop because there'd be no point to it.
I hate sounding like a FICO basher; they have their usefulness. But pretending that it's a precision-guided weapon instead of a nice useful fairly blunt instrument is the thing that's making the investment community run screaming from credit scoring.
I can't wait to hear what the investment community would rather have.
"Did this make anyone else as mad as this made me?"
Yes, it made me angry as well. I read it today and at first I thought "Linda" was trolling but she comes back for more and with more detail as well.
I was somewhat mollified to see the responses to her from some of the other brokers. But if her attitude is common amongst mortgage brokers and they, like her, think that "only a saint" doesn't try tricks like this then I wish that industry festooned in government red-tape for a decade for their nasty little cheating ways.
And per the start of that topic, I'm glad there are rules to clearly state how much the broker is going to make out of the deal. If they are providing a good service they have nothing to fear from that disclosure.
The FICO's a great tool but like many others it has limitations. For instance one of the big issues right now are with first time home buyers. Say you have a young borrower with a great FICO but no depth of credit. If you don't look at the credit report you can overestimate the value of the score.
Take this for the anonymous webposting it is but IBs started kicking loans with high stated incomes and borrower ages under 25 and over 65 over a year ago regardless of FICO.
My guess is that the 720 FICO is more likely to walk way from a upside down load if that is the rational thing to do.
As far as I know, there isn't any correlation between FICO and ruthlessness.
Nonetheless, your point is valid that if you put a high-FICO borrower in a situation in which default is easy and inexpensive, you will find that a high-FICO borrower can be quite ruthless.
Part of the problem is that, since we decreed that anybody can get a loan at (relatively) inexpensive terms, the loss of that pretty FICO just isn't scary enough any more. Or, it wasn't. There are undoubtedly some folks who are going to discover that the subprime industry--not just mortgages, but cars and cards--is in too much disarray right now to accommodate them. If that puts real economic costs back into jingle mail, some people will be less likely to engage in it.
Certainly I would be wary of anyone who counsels walking right now--there's an underlying assumption that credit will remain available (not just expensive) to the FICO-impaired that I don't necessarily share.
I think it's an interesting question as to how the credit rating industry will react to the mountain of foreclosures. One line of thinking (very common but unstated) is that so many people are going into foreclosure that it won't matter that much; "everyone's doing it", so it's "not a big deal." The other line of thinking is that lenders will be so frightened that they won't lend to anybody with any kind of credit blemish at all, period; the "once bitten, twice shy" response. I honestly don't know which direction it will go, but I'm leaning toward "once bitten, twice shy."
FICO and other similar such metrics being backward looking may be great predictors when future conditions follow the pattern of the past but then what happens when things start to get rough. If all you've relied on is that metric then you've got trouble and really, really big trouble if a buyer puts nothing down/has no real cash in the pot equity stake. Even then someone finding themselves upside down with a big mortgage payment in a hole they will never get out of will figure out the meaning of the term sunk costs. There may well be a lot of top quality borrowers who simply have to walk away to save themselves.
If the underlying reality of the situation being measured changes the system doesn't know that.
Metrics Wonk,
If you have insignificant changes to the distribution of product orders in a manufacturing system, alerts go out all over the place to prevent workstation bottlenecks and raw materials outages.
Adapting a tool to a data set for which it was not designed to be sensitive is tricky business.
Tanta,
I hope you get your hands on the Freddie Mac bulletin and toss us the good bits. Thanks again for your great explanation.
"The other line of thinking is that lenders will be so frightened that they won't lend to anybody with any kind of credit blemish at all, period; the "once bitten, twice shy" response."
Past behavior suggests to me that their memories aren't that long. Give it a few years, and the cycle will start back up.
"Did this make anyone else as mad as this made me?"
I notice that posters on the Broker Outpost Mortgage Forums apparently must register and log in. So it would seem that an enterprising FBI agent or other law enforcement person might be able to discover her true identity and take action. That might be possible even without the registration info -- it's a fair bet that her first name actually is Linda, and we know she's been in the business 13 years. And there are probably other posts by her that reveal the state and area where she works, and perhaps which mortgage lenders she works with. Wouldn't it be nice if some clever sleuth were to nail her?
Foreclosures and mortgages have seasonal fluctuations. The change from November to December is 'normal'. I expect 1.2m+ bankruptcies in 2008, and that will be almost right back where we were before the bankruptcy law changed.
Foreclosures: how anyone could buy a home without an attorney is beyond me (I work for an attorney). 40+ pages of legalese is beyond the understanding of just about everyone involved in a real estate closing. But, it happens way too often. I figure more than 50% of all adjustable rate mortgages from the last 4 years will eventually end in foreclosure (the number is probably closer to 75%)
some women in DC made about $20 milliion doing this. Threw herself an $800k wedding at the Mayflower Hotel.
Disappeared a few months ago.
This morning on the way to work I was listening to Psychic Andrea on my favorite radio station and she was talking to a woman who served as strawman on a couple of condo purchases and now she is stuck with them because the actual purchaser changed his mind. Psychic Andrea saw a visit to an attorney's office for the lady but you don't really have to be psychic to predict that.
But what if he gives you a free toaster in the process? That counts as profit right????
Jesus all jokes aside, that really is evil. I'm guessing that the number of people who are in foreclosure and who seek out the freddie mac website is going to be pitifully small sadly.
I'm guessing that the number of people who are in foreclosure and who seek out the freddie mac website is going to be pitifully small sadly.
That's exactly why Freddie gives this to anyone who will take it and show it: churches, community groups, schools, the people next to them on the train.
I'm just doing my part in the "viral marketing" department here.
But then that leaves most(?) facing foreclosure without much better options. They are going to lose away? Sad.
Tanta,
People seem to generally understand that Excel doesn't write your story problems for you, but not so with FICO.
Who do you consider most culpable in accepting FICO as a predictive tool? Were the ratings agencies the dogs who were supposed to bark? Did the original marketing present FICO as a proven and powerful tool?
To bad they didn't do a PSA about 5 years ago warning people about the first batch of fast talking people in suits.
Start something like this: Have you submitted a request for a loan approval. These requests are on file at the credit agencies and mortgage brokers make copies. They then call on you with a large stack of papers and talk about low rates and interest deductions. These brokers get large fees and you are left with an unaffordable loan.
OT - from CNN/Money, 2007 Bankruptcies jump 40%
Bankruptcies jump 40 percent in 2007 - Jan. 3, 2008
Bankruptcies jump 40 percent in 2007
The American Bankruptcy Institute blames the mortgage crisis for heavy debt load, warns that this year could see more bankruptcies.
"NEW YORK (CNNMoney.com) -- The number of Americans filing for consumer bankruptcy increased by nearly 40 percent in 2007, according to the American Bankruptcy Institute.
In a report released Thursday, the ABI said that the number of overall consumer bankruptcy filings reached 801,840 last year, compared to 573,203 in 2006.
"The roughly 40 percent spike in consumer bankruptcies during 2007 presages [an] even higher [number of] filings this year, as the heavy consumer debt load is made worse by the home mortgage crisis," predicted ABI Executive Director Samuel J. Gerdano.
However, the report also showed that the number of bankruptcy filings declined 7.5 percent in December from November. And Chapter 13 filings - those available to individuals with regular income whose debts do not exceed specific amounts - also showed a decline from November to December."
JKB, Exactly what I was thinking while watching.
Who do you consider most culpable in accepting FICO as a predictive tool? Were the ratings agencies the dogs who were supposed to bark? Did the original marketing present FICO as a proven and powerful tool?
That's rather a long story. Also, it's important to distinguish between using FICOs to determine the borrower's eligibility for certain loan terms, and using FICOs to establish basic creditworthiness instead of a full-file review. Those two aren't the same thing.
Originally, FICOs got into the mortgage business via the GSEs. They asked seller/servicers to collect and report them on every loan sold, but they weren't used for any particular purpose by the originator. The idea here was that the GSEs have huge databases of loans, so they could do some "calibrating" of standard loan approvals and FICO scores, to see if they were in fact predictive for mortgage credit (not just consumer credit, which they were designed for).
After several years of data-gathering, the GSEs wrote guidelines to use FICOs to establish due diligence levels (again, not to approve loans with). This meant that if you had FICOs in the lowest bucket, you had to do the most loan-level diligence (verify everything, look at every little detail of the credit report, etc.). In the middle bucket, you could do "normal" diligence. In the top bucket you could do "expedited" review. The idea there was that a high enough FICO creates a presumption of creditworthiness, and so you just looked for anything that might change the presumption. (That's different from the old paranoid way of doing loans, where everybody's presumed to be a deadbeat until you find some reason to qualify them.)
That approach didn't change until the GSEs started expanding their offerings, especially deeper into "flexible" products. Suddenly there were new "options" for borrowers, and so FICOs became a way of measuring eligibility. For instance, the GSEs upped their maximum LTVs for cash-out refis (it had been 75% forever), but only for borrowers who met "enhanced" eligibility criteria. FICO was an easy way to set a separate, overlaid set of criteria.
That's the use of FICOs that the Alt-A and second lien and subprime people picked up. The big problem for them was that there wasn't a bit of a gap between their "standard" loans and their "enhanced" loans, as there was for the GSEs. There was a great yawning chasm. So inevitably FICO started to carry more weight than it was designed to handle.
The rating agencies contributed to the problem because of their preference for quantifiable numbers on data tapes instead of file-level due diligence.
During this whole period, the pricing people also latched onto FICO as a way of making risk-based pricing distinctions. It's hard to do loan-level price add-ons with a computer by using the results of an old-fashioned UW review. But FICO? You can automate the whole process of putting a rate and points and fees on a loan.
Finally, of course, FICOs stopped "calibrating" with LTV (borrower equity) as a default predictor because we went through a boom in which everybody always had equity--even if you didn't have any when you bought, you had some in six months. You can't take performance of loans in a historical period like that and "calibrate" FICOs meaningfully any more than you can calibrate zodiac signs or SAT scores.
But there have, all along, been lenders who doggedly kept using FICOs only in the work-bucketing or eligibility/pricing distinction fashion, not as a core underwriting determinant. You can tell who they are because they're not in BK yet.
No suggestion, apparently, that the lenders should get out in front here and provide some service for their clients. Looks like the real news is that there's a moral hazard in disintermediation.
Did Freddie 'educate' the people two years back about mortgage scams in the same manner? Do you wonder why not?
Most foreclosed houses do not have much equity left after it goes to the bank and gets sold in auction. The money taken away by the foreclosure con-artist hits the pocket book of lender, not the ex-homeowner....now you see why they are concerned.
Did this make anyone else as mad as this made me?
Regarding YSP needing to be disclosed:
"This new YSP disclosure stinks. This may put a damper on bumping the rate up an 1/8 or 1/4 right before closing to get a little extra YSP. They keep coming up with new ways to limit our incomes."
CITI MORTGAGE JUST SENT THIS OUT
So you get to closing and the broker hits you with "Oh you're rate went up" and you're in a real bind and have to close anyways. What they really are doing is pocketing the money. Just wow.
I think they need a museum just for mortgage fraud. Afterall, they just opened one for fraud ....
Museum gives fraud its sordid place in history - USATODAY.com
By the way, what happened to Bacon Dreamz ... did he run away with the Morgage Pig and live in "Hog Heaven"???
.
Did Freddie 'educate' the people two years back about mortgage scams in the same manner? Do you wonder why not?
To be fair, Fannie and Freddie have put money into consumer anti-fraud educational efforts for years. It all just got drowned out by the unending chorus of cheerleaders.
Thank you so much for the history lesson!
Appropos history lessons:
I'd love to have a copy of Freddie Mac Bulletin 96-06. (I think that's the right number. It's the original one one about FICO bucketing.)
It's so old it isn't on All Regs any more, and I no longer own old paper copies of the Selling and Servicing Guides.
If any of you elderly mortgage weenies have a soft copy of it and could send it my way, I'd be grateful. It would be an amusing walk down memory lane.
A few years ago, when prices were going up 50% a year, Fannie and Freddie couldn`t come up with some sort of alert. I guess it was very hard to figure that one out. But somehow they figured out how to get involved in "derivatives" and other weird things. (I said a long time ago on this blog that I expect residential RE to tank at least 60%(Florida) in the next 2-3 years.)
"I WAS ONLY ACCCCTTING!!!
-Master Thespia
"FICO was an easy way ....."
Substitute anything you wish for FICO, and end it with 'to make money', and that's where we stand.
Gold is just the latest example, temporarily, of course.
Not DH
videos gone... darn....
YouTube
- Protect Yourself From Mortgage Fraud
here it is...
Gotta love the subtleties in this one! White guy in a suit scamming women, minorities and the elderly!
A few years ago, when prices were going up 50% a year
Uh, I'm pretty sure I remember being warned about any market in which prices were going up 50% a year.
Dear Tanta how about an old picture of Che? You already have one? OK then.
Dear Tanta how about an old picture of Che? You already have one?
No, but of course I've got my old Marx and Engels Lunchbox from childhood. Will that do?
I'm am wondering if FICO could lose some of its predictive value since it is so well known.
This is 50% of what Suze Orman talks about. It sounds like you could be under a decent amount of financial pressure and fico optimize, as it were. Or you could be in roughly the same position and trash your score simply based on following her strategies.
I work with a complex ontologically based scoring matrix. Nowhere near as well funded as FICO, but in my experience they're great as a tool to inform people who already know wtf they're doing. Once you start hiring people who just live and die by the output file you're about 2-5 years from a shit blizzard (technical term) of problems.
If the underlying reality of the situation being measured changes the system doesn't know that. "Insignificant" variations in the nature of what your data feed reports can produce results that are mind blowingly wrong.
I'm sure the FICO people know this and hire alot of people smarter than me and give them more funding than I can dream of to account for all of this. But in the end there is always going to be the fact that every system is built on a paradigm of historical experience and if reality changes away from that too quickly your system will fail.
Also our sales staff regularly oversells what we actually can do to the point you'd think our existence would negate the necessity for several federal agencies to exist. That's what they're paid for I guess. But I often talk to large customers who think that our engine decides what reality is at any one moment rather than doing a damned good job at quantifying it.
But we do fuck up, more than we ever admit.
So yeah, the fact that FICO is apparently the financial version of St. Peter's ledger rankles me more than a bit.
"you are not going to be in a better situation because some other fast-talking guy in a suit . . . "
Or because some other fast-talking guy in a suit with a beard, a PhD, and the title of "Chairman" asks for some Congressional intervention to get us out of this mess.
My dearest Tanta I can`t believe we have the same lunchbox. You are my hero!(in the old country mine was mostly empty-thanks to Karl Marx)
It sounds like you could be under a decent amount of financial pressure and fico optimize, as it were.
Although one can do short-term gaming of FICOs, I actually think it's less possible or practical than many people do. Like those many people who for only $99.95 will sell you the secret of manipulating your FICO (hint: take the Ginzu knives instead).
The problem, for me, is that we pretend that there's a meaningful distinction in default probability in very small FICO increments. Frankly, the "over 620 OK" "under 620 not OK" is probably quite meaningful. Everything else is, mortgage-wise, closer to talking to yourself.
We pretend that it makes some sort of difference to have a 95% loan with a 720 FICO on one hand and a 95% loan with a 680 FICO on the other. As far as I'm concerned, there's no meaningful difference there. The 680 borrower might be very marginally more likely to get into too much debt and have trouble making payments, but both borrowers can equally easily lose their jobs and both can damned well walk away from upside down loans. As long as they have equity and employment, both of those borrowers will repay you. You might actually make more money, end of the day, on the 680 because while that borrower will end up repaying you, he will also pay you a couple of late fees in there.
Now, a borrower with a 720 today who had a 720 six months ago, on one hand, and a borrower with a 680 today who had a 720 six months ago, might be a different kettle of fish. But we don't look at trends, just today's snapshot. You have to root through the tradelines on the credit report like an old dinosaur underwriter to find trend.
There just isn't any FICO score at which down payment and ability to service debt doesn't matter. It's this last bit of stupidity that is getting the comeuppance, but still there were a lot of distinctions made during the boom that just didn't matter. That's why people are tempted to game their FICOs: they want a 720 instead of a 700 because you get the next price break at 720.
If we made the price-increments bigger we wouldn't have that problem, because you can't usually get enough FICO movement with cosmetic changes in your debts/limits etc. to move your score that much. In other words, if we went back to basing pricing on LTV, DTI etc. and just made the pricing difference under/over 620, the gaming would stop because there'd be no point to it.
I hate sounding like a FICO basher; they have their usefulness. But pretending that it's a precision-guided weapon instead of a nice useful fairly blunt instrument is the thing that's making the investment community run screaming from credit scoring.
I can't wait to hear what the investment community would rather have.
RE: FICO gaming...
I've had a maxim for many years, directly related to Congress, but it's applicable here:
300 million people can outsmart 535 any day of the week.
"Did this make anyone else as mad as this made me?"
Yes, it made me angry as well. I read it today and at first I thought "Linda" was trolling but she comes back for more and with more detail as well.
I was somewhat mollified to see the responses to her from some of the other brokers. But if her attitude is common amongst mortgage brokers and they, like her, think that "only a saint" doesn't try tricks like this then I wish that industry festooned in government red-tape for a decade for their nasty little cheating ways.
And per the start of that topic, I'm glad there are rules to clearly state how much the broker is going to make out of the deal. If they are providing a good service they have nothing to fear from that disclosure.
i found the video heartwarming, especially the music
The FICO's a great tool but like many others it has limitations. For instance one of the big issues right now are with first time home buyers. Say you have a young borrower with a great FICO but no depth of credit. If you don't look at the credit report you can overestimate the value of the score.
Take this for the anonymous webposting it is but IBs started kicking loans with high stated incomes and borrower ages under 25 and over 65 over a year ago regardless of FICO.
My guess is that the 720 FICO is more likely to walk way from a upside down load if that is the rational thing to do.
My guess is that the 720 FICO is more likely to walk way from a upside down load if that is the rational thing to do.
As far as I know, there isn't any correlation between FICO and ruthlessness.
Nonetheless, your point is valid that if you put a high-FICO borrower in a situation in which default is easy and inexpensive, you will find that a high-FICO borrower can be quite ruthless.
Part of the problem is that, since we decreed that anybody can get a loan at (relatively) inexpensive terms, the loss of that pretty FICO just isn't scary enough any more. Or, it wasn't. There are undoubtedly some folks who are going to discover that the subprime industry--not just mortgages, but cars and cards--is in too much disarray right now to accommodate them. If that puts real economic costs back into jingle mail, some people will be less likely to engage in it.
Certainly I would be wary of anyone who counsels walking right now--there's an underlying assumption that credit will remain available (not just expensive) to the FICO-impaired that I don't necessarily share.
I think it's an interesting question as to how the credit rating industry will react to the mountain of foreclosures. One line of thinking (very common but unstated) is that so many people are going into foreclosure that it won't matter that much; "everyone's doing it", so it's "not a big deal." The other line of thinking is that lenders will be so frightened that they won't lend to anybody with any kind of credit blemish at all, period; the "once bitten, twice shy" response. I honestly don't know which direction it will go, but I'm leaning toward "once bitten, twice shy."
FICO and other similar such metrics being backward looking may be great predictors when future conditions follow the pattern of the past but then what happens when things start to get rough. If all you've relied on is that metric then you've got trouble and really, really big trouble if a buyer puts nothing down/has no real cash in the pot equity stake. Even then someone finding themselves upside down with a big mortgage payment in a hole they will never get out of will figure out the meaning of the term sunk costs. There may well be a lot of top quality borrowers who simply have to walk away to save themselves.
If the underlying reality of the situation being measured changes the system doesn't know that.
Metrics Wonk,
If you have insignificant changes to the distribution of product orders in a manufacturing system, alerts go out all over the place to prevent workstation bottlenecks and raw materials outages.
Adapting a tool to a data set for which it was not designed to be sensitive is tricky business.
Tanta,
I hope you get your hands on the Freddie Mac bulletin and toss us the good bits. Thanks again for your great explanation.
there's an underlying assumption that credit will remain available (not just expensive) to the FICO-impaired that I don't necessarily share.
Totally agree, and it's going to be a rude wakeup call to a generation that was born with a silver credit card in it's mouth.
"The other line of thinking is that lenders will be so frightened that they won't lend to anybody with any kind of credit blemish at all, period; the "once bitten, twice shy" response."
Past behavior suggests to me that their memories aren't that long. Give it a few years, and the cycle will start back up.
"Did this make anyone else as mad as this made me?"
I notice that posters on the Broker Outpost Mortgage Forums apparently must register and log in. So it would seem that an enterprising FBI agent or other law enforcement person might be able to discover her true identity and take action. That might be possible even without the registration info -- it's a fair bet that her first name actually is Linda, and we know she's been in the business 13 years. And there are probably other posts by her that reveal the state and area where she works, and perhaps which mortgage lenders she works with. Wouldn't it be nice if some clever sleuth were to nail her?
Foreclosures and mortgages have seasonal fluctuations. The change from November to December is 'normal'. I expect 1.2m+ bankruptcies in 2008, and that will be almost right back where we were before the bankruptcy law changed.
Foreclosures: how anyone could buy a home without an attorney is beyond me (I work for an attorney). 40+ pages of legalese is beyond the understanding of just about everyone involved in a real estate closing. But, it happens way too often. I figure more than 50% of all adjustable rate mortgages from the last 4 years will eventually end in foreclosure (the number is probably closer to 75%)