The question is what the "risk" is. Is it the risk represented by the borrower's past credit performance? In other words, have we given these people credit before--cards, car loans, whatever--and they used too much of it or didn't repay it as agreed?
Or is the "risk" that we cannot measure their riskiness, because we have not granted them enough "starter credit" to let them rack up a history?
Beyond that, the whole "Alt-A" phenomenon is about people who have (putatively) low-risk credit histories, but they are high-risk for mortgage credit because they have to borrow too much in relation to their income.
The whole thing gets further complicated by this apparent reallocation of "spotty" or near-prime credit to subprime. I am not at all confident that the loss in market share of FHA and the growth in subprime over the 2002-2006 period (see spreadsheets I posted in the last two days) has more to do with the inherent riskiness of those loans than it has to do with greater profit opportunities for the originators in conventional (less regulated) subprime than in FHA.
That means that "high risk issue" and "high rate/price issue" have to be unwound to be analyzed. The data I have clearly shows that the average FICOs on subprime have been rising over the last few years. That fact requires some explanation. One possible explanation is that regardless of credit history, you're likely to be subprime if you're trying to buy a home these days, because you're just not likely to be able to afford a home these days.
That changes the whole set of assumptions about who "subprime" borrowers are.
The level of outstanding asset-backed commercial paper fell by $6.8 billion, or 0.8%, to $899 billion in the latest week. The market for such paper has fallen by $284 billion, or 24%, since the credit crunch hit in early August.
And, by the way, if you change your assumptions about who subprime borrowers are, you change your assumptions about workouts. The old "they're just credit abusers anyway" argument gets iffy.
High spreads are the result of either: high RISK factors (credit foremost among them) and/or the possible exisitence of discriminatory pricing. HMDA reporting was created to help detect the latter.
As Tanta pointed out somewhere in there -- ...since the 2002 change to HMDA did not force lenders to collect other data that could account for pricing differences, such as LTV, doc type, FICO, or DTI, you have a set of pricing data but you're back to square one in terms of using it to decide whether this pricing is fair or predatory or discriminatory.
Repackaged? Hell, HMDA reporting would have to be re-engineered entirely to be of any value to its original quest.
One possible explanation is that regardless of credit history, you're likely to be subprime if you're trying to buy a home these days, because you're just not likely to be able to afford a home these days.
Location has to be a big deal here too, right? I mean, wouldn't you expect that the "average" subprime borrower in areas where housing is affordable relative to income (say, Dallas suburbs) would be different then the "average" subprime borrower in Ventura County?
How much sense does it make to look at subprime loans at a national level?
Sorry for the off-topic post, but North Carolina's AG is calling for an investigation of Mozilo's share sales. The SEC is declining comment. Apparantly he's busy selling his 2011 strikes right now - now that's confidence in a Countrywide turnaround (NOT).
Here's another way to look at this, based on data on 2006 subprime first liens (courtesy of UBS).
In the 2006 vintage, 58% of the loans had an original balance of $200,000 or less. 100% of those loans would have fallen into the FHA low-cost county limit (which is 48% of conforming or $200,160).
79% of the loans had an original balance of $300,000 or less. 100% of those loans would have fallen into the FHA high-cost county limit (which is 78% of conforming or $362,790).
With an average FICO of 629 and average CLTV of 87% in that vintage, there's no way a whole bunch of those loans couldn't have qualified for FHA.
isn't it fair, then, to say (repeat?) that what we're seeing is not the result of a "subprime problem" but instead a "credit underwriting standards problem" -- that is, high risk loans became de rigeuer not primarily because of reaching into lower social strata but because of laxity in originator risk management across all social strata? and by extension, across all loan types relative to earlier standards?
it seems to me that wsj (and many others) are continuing to diminish the problem greatly by continuing to allow the word "subprime" to frame the conversation. will we see that sort of admission anytime before recession strikes?
CR, I would like you to look at the euphemism that Michigan is a "one-state recession." What's happening in Michigan is not a recession in any conventional sense. It isn't even a depression.
It's a permanent, irreversible erosion of economic activity, combined with the drag of high debt at all levels. Nobody is offering recipes for curing Michigan's problems, and it's hard to imagine how the state will attract new industry.
It's not "contained" to Michigan, either. It's already spreading to Ohio, Indiana, Illinois, PA, NJ and possibly NY. It's "urban blight" on a regional scale. Really, it's mass economic out-migration.
Okay, you say, so U.S. population and economic activity is migrating South and West. So what?
Massive bankruptcies, defaults, bank failures, and FDIC intervention in the Heartland, combined with soaring taxes (= lower PCE).
Another delightfully long analysis (in blog years) from Tanta.
So, in the bad old unenlightened days of lending, bad risk was tantamount to extra melanin or insufficient testacles. Now the WSJ would have us believe that almost anyone could be a bad risk--that the "problem" crosses all sorts of demographic boundaries.
While I suspect that this is true in some measure, I'd also bet that long-term statistical analysis will show that lenders were in fact preying upon certain classes, and that those classes had darker skin (Black, native American, Hispanic), lower income (see above, in addition to "white trash"), and less traditional family units (single moms).
But that's okay, because healthy markets are so darned good at self-correction. The workouts will come because they HAVE to, and the chastened lenders will reformulate practices to be more sensitive long-run consequences. I'm pretty sure this will happen without any additional government regulation.
When I hear 'subprime', I think a person who has a bad credit record.
What the WSJ seems to have done is redefine all high cost mortgages as subprime, which means that they'll write articles about all kinds of borrowers using language that makes me think of people with bad credit.
That's a real good way to swing public opinion since I'm probably not alone in my preconceptions of what subprime means.
You can't extract deposits from poor people and use them exclusively to fund loans to rich people.
While I don't want to be seen as excusing redlining or racial prejudice, you don't have to be a neocon to read that sentance as equivalent to: "You can't pay interest to poor people and charge it to rich people."
While I don't want to be seen as excusing redlining or racial prejudice, you don't have to be a neocon to read that sentance as equivalent to: "You can't pay interest to poor people and charge it to rich people."
Actually, the proper interpretation would be that "You can't make loans to rich people at 7% while only paying the poor depositors who provided the capital 1.4% and simultaneously offering them loans at 12%"
Speaking of regulation, Treasury has just put out a call for comment on changes to regulation in the banking, insurance and securities industries. This, as we know, is Paulson's favorite pony. All regulation should be reviewed routinely. However, an over-broad, fire-sale approach to regulator change opens the opportunity to, for instance, dump the reporting regulations discussed here.
Not that a guy plucked from the financial industry would do that.
"then you begin to suspect that "subprime" is "loans to naive or desperate borrowers," not this ballyhooed "risk based priced" stuff of recent legend."
When you consider the large number of loans placed by commissioned loan originators, I submit to you that the only reason most loans do not fall into the high rate category is just what you said, "naive or desperate borrowers." All borrowers are not naive or so desperate that they will not shop around. The best restraint against being charged the price at which the YSP flattens out, is for the loan originator to know that you are shopping him.
The objective of most loan originators is to price the loan that will produce the highest YSP along with whatever overt fees he can get, As it happens, this is easiest when you have a naive borrower that trusts what the loan originator says (my job is to shop the market for you and get you the very best loan) or one that is so desperate for the money that they have neither the time, nor feel that they are credit worthy enough, to shop. So many loan originators target borrowers in these categories.
One of the data points omitted from the HMDA report is the race of the loan originator. I will bet that one would be an eye opener, especially with Latino borrowers. MInorities frequently prey on their own, who are especially vulnerable when they cannot read any of the disclosures.
Thanks for the note on the ABCP, those stats aren't up on the Fed site yet.
Interesting to note that is a flat decline amount from previous week, was -$6.1 billion decline now a -$6.8 billion decline (seasonally adjusted number for the -$6.1 billion).
gng, actually the one thing HMDA cannot be beat for is controlling for geography. It is the granddaddy of controlling for geography. All property locations get reported down all the way to census tract level, which is about as micrological as you can get unless you want to go for "block."
So you can in fact very easily control for cost in the HDMA data, and it does in fact show these odd pricing patterns within high-cost areas as well as between cost areas.
Repackaged:. Everything depends on the perspective. The current state could be viewed as:
Perspective A - People with poor credit are defaulting, those slimy bas*****
or
Perspective B - Banks and financial institutions had loose practices and failed to accurately price and assign risk - those slimy bas*****.
Since the market is telling us that recent events cross income, geography, credit histories, I am inclined to think that this is NOT and "Subprime" issue what so ever; it is a "financial mess" due to lax/speculative lending practices.
Just as speculative buying led to the financial disasters in the 1930s so is it that lax/speculative lending has led to this current condition.
European sentiment is that ungrateful US and UK borrowers are defaulting on mortgages and causing losses in the financial markets. This may be true. But at the core it is speculative lending that has rooted it self.
I also think that the billion dollar hiccups of UBS, MER and the like are just the sniffles of a cold which will progress to a financial pandemic.
Sorry to pontificate! And, I am not in the camp that the sky is falling it is just a bit more visible than it was 3 years ago.
On a lighter side, if you live in CA, this is fun:
What about the revelations coming forward that a lot of these sub-prime loans were to people who would/could /should have qualified for prime loans. My gut feeling is that the lending industry realized that there was more money to be had in the sub-prime market and rationalized the use of the product by telling borrowers your can always refinance. So while I can see the possibility that we are all becoming sub-prime candidates because of high LTV due to lack of down payments or low rates to buy the MacMansion, I cant help but feel it was the lenders looking to sack the borrower for higher fees and on top of that being bale to book unrealized profits from the fully adjusted loans. Now there is a racket!!
So you can in fact very easily control for cost in the HDMA data, and it does in fact show these odd pricing patterns within high-cost areas as well as between cost areas.
Wow.
OK then, it's time to let the market sort this out by auctioning off mortgage brokers as indentured servants.
I also took major exception to the line "Subprime mortgages were initially aimed at lower-income consumers with spotty credit." But where Tanta focused on the 'spotty' - it was the suggested 'lower-income' 'aim' that didn't wash with me.
Down where the rubber met the road, '04-06 subprime was a broker-dominated and refinance-oriented business. Those brokers tend to chase after big fish. Which would you rather do? ONE loan for a cardiologist with a bunch of lates thanks to the ex-wife or TEN deals involving city bus drivers with gambling problems, immigrant cleaning ladies with thin credit files, single moms with three jobs, dancers with cash wages and voracious drug habits? - oh, wait, that's alt-a - anyhow, you get the gist. It's not that subprime was ever AIMED at low-income - quite the contrary - it's just that median income of those with impaired credit happens to lower.
In the 2006 vintage, 58% of the loans had an original balance of $200,000 or less. 100% of those loans would have fallen into the FHA low-cost county limit (which is 48% of conforming or $200,160).
79% of the loans had an original balance of $300,000 or less. 100% of those loans would have fallen into the FHA high-cost county limit (which is 78% of conforming or $362,790).
With an average FICO of 629 and average CLTV of 87% in that vintage, there's no way a whole bunch of those loans couldn't have qualified for FHA.
Tanta,
The one thing that is conspicuously absent from your description of these loans is DTI. Did these folks have the DTI to make it into an FHA loan? If not, perhaps it does all come down to home price inflation and wage stagnation, and "we are all subprime now".
FKA, I think it's a lot of things at once. The search for "the explanation" drives me up the wall. Many things are going on at once, and the urge to collapse them all into "bites" is a huge problem. Among other things, you get crappy regulatory responses to crappy analysis of the problem.
So you can believe that "predatory steering" has gone on--that people are getting subprime loans who should have gotten prime loans--at the same time that you can believe that putting them in prime loans wouldn't have helped them any if homes are just unaffordable under any terms.
I'm not personally real enthusiastic about putting people in homes if they cost too much and they are not affordable in the long term. However, if we're going to do that--and we've been livin' in this "ownership society" shit for years saying that this is a good thing--then by god why are some people paying higher interest rates to mess up their own futures via homeownership than others? If I'm going to let Skippy and Buffy impoverish themselves on homeownership at 6.25%, then why am I justified in only letting Juan and Yolanda do that if they pay 8.50% for the same damned loan? It makes sense to make the most vulnerable part of the population go to financial hell faster than others?
I have spent way too much of my time over the years battling the rear-guard who want to claim that "nothing can be done" for these borrowers in predatory loans because they couldn't afford homeownership whatever you do for them. Now that we're finding out that the only reason the "prime" crowd could afford to buy homes is because they sold their outrageously overpriced existing home to some hapless FTHB with a toxic mortgage, some folks are worked up over finally wanting to do something.
And what do they want to do? Raise loan limits. In the face of piles o' data showing that nobody can afford those bigger loans without "toxifying" the terms (teasers, IO, neg am, 50-year terms, etc.). In other words, the apparent solution is really to "make us all subprime."
If this ain't "just a subprime problem," then an entire debt-based economy in which even the middle and upper middle class cannot afford homes given RE inflation and wage stagnation is suddenly in question. The last thing certain vested interests want to hear is that, basically, "we are all subprime now."
Kinda gives you the picture that we're all holding on by our fingernails?
The one thing that is conspicuously absent from your description of these loans is DTI.
Funny you should ask. Average DTI on that vintage is 41%. 41% is . . . the FHA maximum!
But of course, that DTI is based on the subprime interest rates, which in many cases were a lot higher than the FHA rate would have been. So you can't really be sure that the DTI wouldn't have been lower in FHA.
OT: A friend of mine is looking to refi out of his ARM into a fixed loan. This is the exact text that his Countrywide loan agent sent him in an email:
"your credit is excellent so I will not need documentation from you it is called Fast and Easy."
It would appear to me that Countrywide is still doing no doc loans even though I thought most of those products were scrapped because they couldn't be sold in the secondary market. He was quoted a non-conforming rate of 7%. How are they still able to do these types of loans?
Alex, HMDA "got through" in 1975. The updated law "got through" in 1989, when financial institutions were blowing up all over the place and they weren't in much of a position to lobby against more reporting regulation. The 1989 law gave the Fed the authority to add reporting requirements without further congressional action.
And it took the Fed until 2002 to finally add just simple price data.
You notice they didn't add the other data, like FICO or LTV or what have you, that would have really made sense of that price data.
So I'd not be inclined to congratulate the 2002 Fed leadership in progressive regulation.
"And what do they want to do? Raise loan limits. In the face of piles o' data showing that nobody can afford those bigger loans without "toxifying" the terms (teasers, IO, neg am, 50-year terms, etc.). In other words, the apparent solution is really to "make us all subprime."
Yes, because that is the only way for a stagnant wage to afford a house. Once again, when someone buys a house at 10x their annual income, they are speculating on future appreciation whether they realize it or not.
Everyone wants to dance around the elephant in the room: House Prices in Bubble Areas have to CRASH!!! Period. End of discussion.
When an asset gets inflated via a credit bubble then eventually debt services exceeds ability to pay and the debtor effectively becomes "Subprime".
You folks rock! This is the first time I have posted in the comments, but I have read this blog and most every entry within it for the last two years. I am a professor and sociologist at a large state university, and today, I will be distributing this essay on the HMDA data in my social stratification course. Many of my students come from families, or are themselves (the case of a single mom in my course) who are now losing their homes due to resetting rates and foreclosures. Most of these folks, mostly well-meaning working- and service-class parents, simply didn't have the financial knowledge and math skills they really needed when they decided to finance their homes. In the process of explaining to college seniors that the dynamics of race, gender and class play an important role in creating inequality in the financial markets as well as the efforts of political institutions to make these markets more accountable for this (e.g., HMDA,CRA, Tanta's post will work perfectly to reveal the dynamics of social stratification.
Thank you very much Tanta and CR for your dedication to educating the public and revealing the chicanery that goes on in the mortgage and lending markets, and in the newsrooms of some mainstream media. Ok, it's not always chicanery, sometimes its just laziness and stupidity, but I can always count on getting clarity at Calculated Risk. I am looking forward to sharing this post with colleagues, neighbors, friends, and in a political action committee to which I belong. Maybe, we can begin to build some political pressure to require lenders to collect data on LTV, doc type, FICO in California and find out if high risk and high price actually do cohere in the manner the Wall Street Journal assumes that they do.
From the Article: (i.e., those deposits need to be "reinvested" in the neighbhorhood they came from in the form of loans, not just mortgage loans, to that neighborhood. You can't extract deposits from poor people and use them exclusively to fund loans to rich people.)
Thought: Except if those poor people live in China. Then they can lend the people in the USA over almost $200 billion per year. Go trade deficit!!!
From 2007-2009, 2.3 million adjustable rate mortgages are expected to reset, nearly a quarter of which will be at risk of foreclosure. FHA believes that FHASecure will be able to assist approximately 80,000 borrowers who are delinquent due to their loan resetting, as well as 160,000 borrowers who are facing reset but are still current on their mortgage. By offering FHASecure and other refinancing programs, FHA could help 240,000 borrowers facing reset.
Christopher Carrington - good of you to bring your lurking to an end and join the commentary fun. However, I am not sure I understand what you are after...
require lenders to collect data on LTV, doc type, FICO in California and find out if high risk and high price actually do cohere in the manner the Wall Street Journal assumes
I assure you lenders already COLLECT data on all these attributes and more. Maybe you meant "report it to the FFIEC"?
Even if they were to do so, I can already promise you that this would support the basic premise that price is positively correlated with LTV and negatively with FICO and DOC level.
The sloppy assumption in the WSJ article in question is in its implication that the "high cost loans" (as per HMDA definition) are synonymous with "subprime" loans.
If your concern (I'm guessing here) is that protected classes are receiving disparate pricing, you need to first hold all the risk factors constant. Once you consider do so, I think you would be surprised how 'fair' the lenders are. Mostly because they all use AUSes now (much to Tanta's chagrin) and an AUS doesn't know your ethnic background, sex, etc.
Dr. Carrington, thank you for the comment; you warm my heart.
Blog posts hit the college curriculum. Well hot damn. Take that, Wall Street Journal!
Seriously, we owe you one for helping us prove that "citizen journalism" can make some sort of a difference to anyone other than the short sellers or "bitter renters." I hope you encourage your students to come participate in the comments here. Too many of us are old farts; we could use some new perspectives.
The Boston Fed's been researching sub-primes (see Second Question in the speech) and found: Guess what! A lot of sub-prime borrowers may well qualify for prime loans.
Right. And even when they do, you have 40% using stated income, plus most of them being qualified at a teaser rate. So DTI doesn't mean much, which is why I didn't focus on it to start with.
Schaps Parlor, I do think AUSes are less discriminatory than human underwriters. I remember the bad old days quite vividly, thanks.
That said. First, there's "prequalification," which is is a big knotty issue. If you "screen out" minorities to start with, your AUS gets a nice homogenous pool of borrowers among which to make distinctions. Wow! It doesn't discriminate!
Second, there's AUS and then there's AUS. The machines approve or deny loans. The GSE ones are better at having the capacity to put the borrower in the right product than some of the others, but none of them to my knowledge do the pricing. That was my whole point with FHA loans: if you follow the rules on allowable closing costs that HUD lays down, and you price those loans given Ginnie Mae rules for note rate bands in pools, you just can't do much gouging. But if you're telling me that subprime AUS always put the borrowers in the best rate they can qualify for? Nah, don't buy it.
Third, the big problem is depository lenders just exiting some market areas, and leaving them to the subprime sharks. Suburban borrowers can find a HMDA-reporting depository or big mortgage company lender office on just about any block. Urban and rural and low-income areas generally find nothing but mortgage brokers. With brokers doing the "pre-screening" and the pricing, the AUS is just ratifying what the brokers do.
I do agree that raw race-based discrimination in loan approvals are much less of a problem than they used to be. But I do not agree that pricing is a level playing field across racial and economic levels.
You can automate underwriters, but until you automate loan officers and brokers, somebody with not just biases but a financial incentive in wringing the most out of the least sophisticated is mediating the decision.
Let me say something about screening, just to illustrate how actually very good intentions can harm certain borrowers.
Back in the days when I worked for a retail lender, we would charge a non-refundable application fee. That kept the non-serious folks out of the pipeline, and actually kept costs down to everyone (since we were compensated for the time and energy it took to do the analysis needed to legitimately turn down bad credit risks.) The application fee was equal to the cost of the credit report and appraisal.
Well, that's a fair amount of money for a low-income person to blow if they have no hope of qualifying for a loan. A lot of us used to go after the loan officers like the wrath of God because they were just extracting fees and wasting the time and hopes of people who should have been sent to a homebuyer counseling program and told to wait.
Well, that backfired for real, because it was like giving the LOs a license to screen people, and as Shnaps indicated above, when times were fat and there were plenty of doctor loans to go around, the LOs didn't want to screw with little bitty FHA loans to working people.
Nowadays, of course, "competitive" pressures have eliminated the upfront nonrefundable application fee. The result of which is that if you are misguided enough to apply for a loan, no one will let you get away without one because you must close for them to get compensation. You'll get put in some evil toxic loan if that's what it takes to qualify you.
So these things are not simple, and they have a lot to do with business structures and practices that don't have an obvious connection to race or class (that's the "disparate impact" school of thought, that considers how business practices that are not intentionally discriminatory can have disparate impacts on different kinds of borrowers).
1) "Magisterial" is the word that comes to mind to describe Tanta today.
2) Wow, an actual falsifiable hypothesis? Karl Popper would be very proud of Tanta!!
3) I am sure I have missed a thread or two, but speaking hypotheses, has FICO as a measure of outcome (servicing mortgage debt) been rigorously tested? To what extent is FICO a proprietary black box? And has there been grade inflation (has the method changed, is it grading on a curve?) Finally, has the predictive value of a particular score changed ovewr time? IOW, does a 680 from 2001 show the same behavior as a 680 from 2006 or is it too early to tell?
Tanta, thanks for bringing up some interesting considerations.
Sometimes I get so exasperated trying to explain that an AUS is egalitarian by design, I don't step back and ponder other systemic factors that may be surrounding the AUS.
I'm all for this blog as required reading. In fact, the good Dr. Carrington should invite you out for a guest lecture! With a paper bag over your head to maintain anonymity, of course.
Golly, David, you really want to go for the extra credit, don't you?
I shall try to write something on recent indications that FICOs aren't as well calibrated with mortgages as we have all been told.
For now, let me just say that there was, originally, a serious effort to do that. The GSEs got interested in FICOs in the mid-90s, and they started by just requiring them to be reported on loan deliveries. You didn't use them to make cutoffs or price loans or anything, you just collected the data and passed it on.
That allowed the GSEs, who have these big honkin' geographically diverse databases, to do some "calibrating" of FICO with human underwriting, in terms of qualification, and FICO with subsequent servicing history, in terms of performance.
The first result of that research, actually, came in GSE recommendations that FICO be used to set level of review for loan files. You didn't make the decision based on the FICO, but lower FICO loans had to be subjected to more detailed review than higher FICO loans. As an example, if the FICO was 720 or higher, all you had to look at in detail was recent (last 12-24 months) of credit history to make sure it wasn't trending down. If the FICO was less than 660, you really needed to wade through the whole picture, and get written explanations & more documentation from the borrower to account for the problems.
It was not until years after that that the GSEs started using FICO cutoffs for certain kinds of eligiblity. But they have never used FICOs in their own underwriting. Their AUSs do their own analysis of the full credit report file. People don't realize this: the AUS gets an electronic file of the whole credit report, not just a 3-digit FICO code. But when loans are manually underwritten, the agencies require certain baseline FICOs before you can offer certain kinds of maximum financing or other risk factors.
Eventually it gets hard to "recalibrate" FICOs to mortgage performance, because they are used to sort borrowers into various products (prime, alt-a, subprime), and so then you get FICO bands overlaying product bands, and then you can't tell which is driving what (the FICO or the loan terms themselves). It takes a HUGE database of loans to get significant results.
HUGE database. We're lookin' at you, rating agencies. If you expect individual lenders to do this analysis, it will be skewed by the individual lender's product line or geographical base.
The basic problem with sub-prime was that the lenders developed a bias based on "collateral dependency" which regulators consider to be unsound lending practices.
The belief and bias was so strong that "you can always sell or refinance". We see where that led to.
Yo Tanta. Much better. Our risk now is going to far the other way. The market is correcting itself. Lenders now have fear. But too much fear is a bad thing as is too much greed.
The over pricing loans is increasing because many people are applying for loans. So the owner was increasing the interest so that he/she could have bigger profit. And the person who are engaged with this matter are very rich with income.
So, its not a subprime issue, it is a high risk credit issue...this has to be repackaged for historical review
The last thing certain vested interests want to hear is that, basically, "we are all subprime now."
At least then, when they say the problem is contained to subprime, they'd be correct.
The question is what the "risk" is. Is it the risk represented by the borrower's past credit performance? In other words, have we given these people credit before--cards, car loans, whatever--and they used too much of it or didn't repay it as agreed?
Or is the "risk" that we cannot measure their riskiness, because we have not granted them enough "starter credit" to let them rack up a history?
Beyond that, the whole "Alt-A" phenomenon is about people who have (putatively) low-risk credit histories, but they are high-risk for mortgage credit because they have to borrow too much in relation to their income.
The whole thing gets further complicated by this apparent reallocation of "spotty" or near-prime credit to subprime. I am not at all confident that the loss in market share of FHA and the growth in subprime over the 2002-2006 period (see spreadsheets I posted in the last two days) has more to do with the inherent riskiness of those loans than it has to do with greater profit opportunities for the originators in conventional (less regulated) subprime than in FHA.
That means that "high risk issue" and "high rate/price issue" have to be unwound to be analyzed. The data I have clearly shows that the average FICOs on subprime have been rising over the last few years. That fact requires some explanation. One possible explanation is that regardless of credit history, you're likely to be subprime if you're trying to buy a home these days, because you're just not likely to be able to afford a home these days.
That changes the whole set of assumptions about who "subprime" borrowers are.
Asset-backed paper falls for ninth straight week
Asset-backed paper falls for ninth straight week - MarketWatch
The level of outstanding asset-backed commercial paper fell by $6.8 billion, or 0.8%, to $899 billion in the latest week. The market for such paper has fallen by $284 billion, or 24%, since the credit crunch hit in early August.
And, by the way, if you change your assumptions about who subprime borrowers are, you change your assumptions about workouts. The old "they're just credit abusers anyway" argument gets iffy.
Barley - you didn't RTFP, did you?
High spreads are the result of either: high RISK factors (credit foremost among them) and/or the possible exisitence of discriminatory pricing. HMDA reporting was created to help detect the latter.
As Tanta pointed out somewhere in there -- ...since the 2002 change to HMDA did not force lenders to collect other data that could account for pricing differences, such as LTV, doc type, FICO, or DTI, you have a set of pricing data but you're back to square one in terms of using it to decide whether this pricing is fair or predatory or discriminatory.
Repackaged? Hell, HMDA reporting would have to be re-engineered entirely to be of any value to its original quest.
One possible explanation is that regardless of credit history, you're likely to be subprime if you're trying to buy a home these days, because you're just not likely to be able to afford a home these days.
Location has to be a big deal here too, right? I mean, wouldn't you expect that the "average" subprime borrower in areas where housing is affordable relative to income (say, Dallas suburbs) would be different then the "average" subprime borrower in Ventura County?
How much sense does it make to look at subprime loans at a national level?
Sorry for the off-topic post, but North Carolina's AG is calling for an investigation of Mozilo's share sales. The SEC is declining comment. Apparantly he's busy selling his 2011 strikes right now - now that's confidence in a Countrywide turnaround (NOT).
Here's another way to look at this, based on data on 2006 subprime first liens (courtesy of UBS).
In the 2006 vintage, 58% of the loans had an original balance of $200,000 or less. 100% of those loans would have fallen into the FHA low-cost county limit (which is 48% of conforming or $200,160).
79% of the loans had an original balance of $300,000 or less. 100% of those loans would have fallen into the FHA high-cost county limit (which is 78% of conforming or $362,790).
With an average FICO of 629 and average CLTV of 87% in that vintage, there's no way a whole bunch of those loans couldn't have qualified for FHA.
isn't it fair, then, to say (repeat?) that what we're seeing is not the result of a "subprime problem" but instead a "credit underwriting standards problem" -- that is, high risk loans became de rigeuer not primarily because of reaching into lower social strata but because of laxity in originator risk management across all social strata? and by extension, across all loan types relative to earlier standards?
it seems to me that wsj (and many others) are continuing to diminish the problem greatly by continuing to allow the word "subprime" to frame the conversation. will we see that sort of admission anytime before recession strikes?
OT
CR, I would like you to look at the euphemism that Michigan is a "one-state recession." What's happening in Michigan is not a recession in any conventional sense. It isn't even a depression.
It's a permanent, irreversible erosion of economic activity, combined with the drag of high debt at all levels. Nobody is offering recipes for curing Michigan's problems, and it's hard to imagine how the state will attract new industry.
It's not "contained" to Michigan, either. It's already spreading to Ohio, Indiana, Illinois, PA, NJ and possibly NY. It's "urban blight" on a regional scale. Really, it's mass economic out-migration.
Okay, you say, so U.S. population and economic activity is migrating South and West. So what?
Massive bankruptcies, defaults, bank failures, and FDIC intervention in the Heartland, combined with soaring taxes (= lower PCE).
Another delightfully long analysis (in blog years) from Tanta.
So, in the bad old unenlightened days of lending, bad risk was tantamount to extra melanin or insufficient testacles. Now the WSJ would have us believe that almost anyone could be a bad risk--that the "problem" crosses all sorts of demographic boundaries.
While I suspect that this is true in some measure, I'd also bet that long-term statistical analysis will show that lenders were in fact preying upon certain classes, and that those classes had darker skin (Black, native American, Hispanic), lower income (see above, in addition to "white trash"), and less traditional family units (single moms).
But that's okay, because healthy markets are so darned good at self-correction. The workouts will come because they HAVE to, and the chastened lenders will reformulate practices to be more sensitive long-run consequences. I'm pretty sure this will happen without any additional government regulation.
Oh, no I'm not.
When I hear 'subprime', I think a person who has a bad credit record.
What the WSJ seems to have done is redefine all high cost mortgages as subprime, which means that they'll write articles about all kinds of borrowers using language that makes me think of people with bad credit.
That's a real good way to swing public opinion since I'm probably not alone in my preconceptions of what subprime means.
You can't extract deposits from poor people and use them exclusively to fund loans to rich people.
While I don't want to be seen as excusing redlining or racial prejudice, you don't have to be a neocon to read that sentance as equivalent to: "You can't pay interest to poor people and charge it to rich people."
Jim a:
While I don't want to be seen as excusing redlining or racial prejudice, you don't have to be a neocon to read that sentance as equivalent to: "You can't pay interest to poor people and charge it to rich people."
Actually, the proper interpretation would be that "You can't make loans to rich people at 7% while only paying the poor depositors who provided the capital 1.4% and simultaneously offering them loans at 12%"
Speaking of regulation, Treasury has just put out a call for comment on changes to regulation in the banking, insurance and securities industries. This, as we know, is Paulson's favorite pony. All regulation should be reviewed routinely. However, an over-broad, fire-sale approach to regulator change opens the opportunity to, for instance, dump the reporting regulations discussed here.
Not that a guy plucked from the financial industry would do that.
"Greatest financial scam in history". Ohio Attorney General Mark Dan is on CNBC weighing in right now on the rating agencies and Wall Street.
kurtyboy: well said, point taken.
Mark Dan says civil cases vs. Wall Street dealers are in the works, in addition to criminal indictments against mortgage brokers.
"then you begin to suspect that "subprime" is "loans to naive or desperate borrowers," not this ballyhooed "risk based priced" stuff of recent legend."
When you consider the large number of loans placed by commissioned loan originators, I submit to you that the only reason most loans do not fall into the high rate category is just what you said, "naive or desperate borrowers." All borrowers are not naive or so desperate that they will not shop around. The best restraint against being charged the price at which the YSP flattens out, is for the loan originator to know that you are shopping him.
The objective of most loan originators is to price the loan that will produce the highest YSP along with whatever overt fees he can get, As it happens, this is easiest when you have a naive borrower that trusts what the loan originator says (my job is to shop the market for you and get you the very best loan) or one that is so desperate for the money that they have neither the time, nor feel that they are credit worthy enough, to shop. So many loan originators target borrowers in these categories.
One of the data points omitted from the HMDA report is the race of the loan originator. I will bet that one would be an eye opener, especially with Latino borrowers. MInorities frequently prey on their own, who are especially vulnerable when they cannot read any of the disclosures.
REBear,
Thanks for the note on the ABCP, those stats aren't up on the Fed site yet.
Interesting to note that is a flat decline amount from previous week, was -$6.1 billion decline now a -$6.8 billion decline (seasonally adjusted number for the -$6.1 billion).
(which is 78% of conforming or $362,790).
Yes, lysdexia stuck Tanta again. Graduates of the third grade will notice that I meant 87%, not 78%.
Dear Tanta
Thanks...very insightful even for my "reading and learning" mind.
BTW, Berry of Bloomberg also had something on this topic today:
There May Be an Out for Some Subprime Borrowers: John M. Berry - Bloomberg.com
Hope all is well with you.
Best regards,
gng, actually the one thing HMDA cannot be beat for is controlling for geography. It is the granddaddy of controlling for geography. All property locations get reported down all the way to census tract level, which is about as micrological as you can get unless you want to go for "block."
So you can in fact very easily control for cost in the HDMA data, and it does in fact show these odd pricing patterns within high-cost areas as well as between cost areas.
Repackaged:. Everything depends on the perspective. The current state could be viewed as:
Perspective A - People with poor credit are defaulting, those slimy bas*****
or
Perspective B - Banks and financial institutions had loose practices and failed to accurately price and assign risk - those slimy bas*****.
Since the market is telling us that recent events cross income, geography, credit histories, I am inclined to think that this is NOT and "Subprime" issue what so ever; it is a "financial mess" due to lax/speculative lending practices.
Just as speculative buying led to the financial disasters in the 1930s so is it that lax/speculative lending has led to this current condition.
European sentiment is that ungrateful US and UK borrowers are defaulting on mortgages and causing losses in the financial markets. This may be true. But at the core it is speculative lending that has rooted it self.
I also think that the billion dollar hiccups of UBS, MER and the like are just the sniffles of a cold which will progress to a financial pandemic.
Sorry to pontificate! And, I am not in the camp that the sky is falling it is just a bit more visible than it was 3 years ago.
On a lighter side, if you live in CA, this is fun:
ForeclosureRadar
If Mark Dan and the State of Ohio want to put Moody's out of business, I believe they have the power.
Of course, their main interest isn't that, but rather to recover losses suffered by the People of Ohio.
I believe they will recover billions. The damage this scam has inflicted on Ohio is monumental.
tanta,
What about the revelations coming forward that a lot of these sub-prime loans were to people who would/could /should have qualified for prime loans. My gut feeling is that the lending industry realized that there was more money to be had in the sub-prime market and rationalized the use of the product by telling borrowers your can always refinance. So while I can see the possibility that we are all becoming sub-prime candidates because of high LTV due to lack of down payments or low rates to buy the MacMansion, I cant help but feel it was the lenders looking to sack the borrower for higher fees and on top of that being bale to book unrealized profits from the fully adjusted loans. Now there is a racket!!
So you can in fact very easily control for cost in the HDMA data, and it does in fact show these odd pricing patterns within high-cost areas as well as between cost areas.
Wow.
OK then, it's time to let the market sort this out by auctioning off mortgage brokers as indentured servants.
I also took major exception to the line "Subprime mortgages were initially aimed at lower-income consumers with spotty credit." But where Tanta focused on the 'spotty' - it was the suggested 'lower-income' 'aim' that didn't wash with me.
Down where the rubber met the road, '04-06 subprime was a broker-dominated and refinance-oriented business. Those brokers tend to chase after big fish. Which would you rather do? ONE loan for a cardiologist with a bunch of lates thanks to the ex-wife or TEN deals involving city bus drivers with gambling problems, immigrant cleaning ladies with thin credit files, single moms with three jobs, dancers with cash wages and voracious drug habits? - oh, wait, that's alt-a - anyhow, you get the gist. It's not that subprime was ever AIMED at low-income - quite the contrary - it's just that median income of those with impaired credit happens to lower.
Tanta,
Simply put: you're a treasure. Thanks.
Omark
In the 2006 vintage, 58% of the loans had an original balance of $200,000 or less. 100% of those loans would have fallen into the FHA low-cost county limit (which is 48% of conforming or $200,160).
79% of the loans had an original balance of $300,000 or less. 100% of those loans would have fallen into the FHA high-cost county limit (which is 78% of conforming or $362,790).
With an average FICO of 629 and average CLTV of 87% in that vintage, there's no way a whole bunch of those loans couldn't have qualified for FHA.
Tanta,
The one thing that is conspicuously absent from your description of these loans is DTI. Did these folks have the DTI to make it into an FHA loan? If not, perhaps it does all come down to home price inflation and wage stagnation, and "we are all subprime now".
FKA, I think it's a lot of things at once. The search for "the explanation" drives me up the wall. Many things are going on at once, and the urge to collapse them all into "bites" is a huge problem. Among other things, you get crappy regulatory responses to crappy analysis of the problem.
So you can believe that "predatory steering" has gone on--that people are getting subprime loans who should have gotten prime loans--at the same time that you can believe that putting them in prime loans wouldn't have helped them any if homes are just unaffordable under any terms.
I'm not personally real enthusiastic about putting people in homes if they cost too much and they are not affordable in the long term. However, if we're going to do that--and we've been livin' in this "ownership society" shit for years saying that this is a good thing--then by god why are some people paying higher interest rates to mess up their own futures via homeownership than others? If I'm going to let Skippy and Buffy impoverish themselves on homeownership at 6.25%, then why am I justified in only letting Juan and Yolanda do that if they pay 8.50% for the same damned loan? It makes sense to make the most vulnerable part of the population go to financial hell faster than others?
I have spent way too much of my time over the years battling the rear-guard who want to claim that "nothing can be done" for these borrowers in predatory loans because they couldn't afford homeownership whatever you do for them. Now that we're finding out that the only reason the "prime" crowd could afford to buy homes is because they sold their outrageously overpriced existing home to some hapless FTHB with a toxic mortgage, some folks are worked up over finally wanting to do something.
And what do they want to do? Raise loan limits. In the face of piles o' data showing that nobody can afford those bigger loans without "toxifying" the terms (teasers, IO, neg am, 50-year terms, etc.). In other words, the apparent solution is really to "make us all subprime."
If this ain't "just a subprime problem," then an entire debt-based economy in which even the middle and upper middle class cannot afford homes given RE inflation and wage stagnation is suddenly in question. The last thing certain vested interests want to hear is that, basically, "we are all subprime now."
Kinda gives you the picture that we're all holding on by our fingernails?
The one thing that is conspicuously absent from your description of these loans is DTI.
Funny you should ask. Average DTI on that vintage is 41%. 41% is . . . the FHA maximum!
But of course, that DTI is based on the subprime interest rates, which in many cases were a lot higher than the FHA rate would have been. So you can't really be sure that the DTI wouldn't have been lower in FHA.
OT: A friend of mine is looking to refi out of his ARM into a fixed loan. This is the exact text that his Countrywide loan agent sent him in an email:
"your credit is excellent so I will not need documentation from you it is called Fast and Easy."
It would appear to me that Countrywide is still doing no doc loans even though I thought most of those products were scrapped because they couldn't be sold in the secondary market. He was quoted a non-conforming rate of 7%. How are they still able to do these types of loans?
Tanta and Cr:
I am humbled and bow to your knowledge. Cheers.
Surprising that legislation as progressive-sounding as the HMDA was able to get through in 2002.
Alex, HMDA "got through" in 1975. The updated law "got through" in 1989, when financial institutions were blowing up all over the place and they weren't in much of a position to lobby against more reporting regulation. The 1989 law gave the Fed the authority to add reporting requirements without further congressional action.
And it took the Fed until 2002 to finally add just simple price data.
You notice they didn't add the other data, like FICO or LTV or what have you, that would have really made sense of that price data.
So I'd not be inclined to congratulate the 2002 Fed leadership in progressive regulation.
"And what do they want to do? Raise loan limits. In the face of piles o' data showing that nobody can afford those bigger loans without "toxifying" the terms (teasers, IO, neg am, 50-year terms, etc.). In other words, the apparent solution is really to "make us all subprime."
Yes, because that is the only way for a stagnant wage to afford a house. Once again, when someone buys a house at 10x their annual income, they are speculating on future appreciation whether they realize it or not.
Everyone wants to dance around the elephant in the room: House Prices in Bubble Areas have to CRASH!!! Period. End of discussion.
When an asset gets inflated via a credit bubble then eventually debt services exceeds ability to pay and the debtor effectively becomes "Subprime".
Tanta and CR,
You folks rock! This is the first time I have posted in the comments, but I have read this blog and most every entry within it for the last two years. I am a professor and sociologist at a large state university, and today, I will be distributing this essay on the HMDA data in my social stratification course. Many of my students come from families, or are themselves (the case of a single mom in my course) who are now losing their homes due to resetting rates and foreclosures. Most of these folks, mostly well-meaning working- and service-class parents, simply didn't have the financial knowledge and math skills they really needed when they decided to finance their homes. In the process of explaining to college seniors that the dynamics of race, gender and class play an important role in creating inequality in the financial markets as well as the efforts of political institutions to make these markets more accountable for this (e.g., HMDA,CRA, Tanta's post will work perfectly to reveal the dynamics of social stratification.
Thank you very much Tanta and CR for your dedication to educating the public and revealing the chicanery that goes on in the mortgage and lending markets, and in the newsrooms of some mainstream media. Ok, it's not always chicanery, sometimes its just laziness and stupidity, but I can always count on getting clarity at Calculated Risk. I am looking forward to sharing this post with colleagues, neighbors, friends, and in a political action committee to which I belong. Maybe, we can begin to build some political pressure to require lenders to collect data on LTV, doc type, FICO in California and find out if high risk and high price actually do cohere in the manner the Wall Street Journal assumes that they do.
franz
from sf?
From the Article: (i.e., those deposits need to be "reinvested" in the neighbhorhood they came from in the form of loans, not just mortgage loans, to that neighborhood. You can't extract deposits from poor people and use them exclusively to fund loans to rich people.)
Thought: Except if those poor people live in China. Then they can lend the people in the USA over almost $200 billion per year.
Go trade deficit!!!
FHA says:
From 2007-2009, 2.3 million adjustable rate mortgages are expected to reset, nearly a quarter of which will be at risk of foreclosure. FHA believes that FHASecure will be able to assist approximately 80,000 borrowers who are delinquent due to their loan resetting, as well as 160,000 borrowers who are facing reset but are still current on their mortgage. By offering FHASecure and other refinancing programs, FHA could help 240,000 borrowers facing reset.
Average DTI on that vintage is 41%
a lot of loans don't report DTI, though, right?
Christopher Carrington - good of you to bring your lurking to an end and join the commentary fun. However, I am not sure I understand what you are after...
require lenders to collect data on LTV, doc type, FICO in California and find out if high risk and high price actually do cohere in the manner the Wall Street Journal assumes
I assure you lenders already COLLECT data on all these attributes and more. Maybe you meant "report it to the FFIEC"?
Even if they were to do so, I can already promise you that this would support the basic premise that price is positively correlated with LTV and negatively with FICO and DOC level.
The sloppy assumption in the WSJ article in question is in its implication that the "high cost loans" (as per HMDA definition) are synonymous with "subprime" loans.
If your concern (I'm guessing here) is that protected classes are receiving disparate pricing, you need to first hold all the risk factors constant. Once you consider do so, I think you would be surprised how 'fair' the lenders are. Mostly because they all use AUSes now (much to Tanta's chagrin) and an AUS doesn't know your ethnic background, sex, etc.
Dr. Carrington, thank you for the comment; you warm my heart.
Blog posts hit the college curriculum. Well hot damn. Take that, Wall Street Journal!
Seriously, we owe you one for helping us prove that "citizen journalism" can make some sort of a difference to anyone other than the short sellers or "bitter renters." I hope you encourage your students to come participate in the comments here. Too many of us are old farts; we could use some new perspectives.
Hi Tanta,
Did you see this: FRBB: Speeches- Recent Developments in Real Estate, Financial Markets, and the Economy
The Boston Fed's been researching sub-primes (see Second Question in the speech) and found: Guess what! A lot of sub-prime borrowers may well qualify for prime loans.
Thank you for all your great posts!
a lot of loans don't report DTI, though, right?
Right. And even when they do, you have 40% using stated income, plus most of them being qualified at a teaser rate. So DTI doesn't mean much, which is why I didn't focus on it to start with.
Schaps Parlor, I do think AUSes are less discriminatory than human underwriters. I remember the bad old days quite vividly, thanks.
That said. First, there's "prequalification," which is is a big knotty issue. If you "screen out" minorities to start with, your AUS gets a nice homogenous pool of borrowers among which to make distinctions. Wow! It doesn't discriminate!
Second, there's AUS and then there's AUS. The machines approve or deny loans. The GSE ones are better at having the capacity to put the borrower in the right product than some of the others, but none of them to my knowledge do the pricing. That was my whole point with FHA loans: if you follow the rules on allowable closing costs that HUD lays down, and you price those loans given Ginnie Mae rules for note rate bands in pools, you just can't do much gouging. But if you're telling me that subprime AUS always put the borrowers in the best rate they can qualify for? Nah, don't buy it.
Third, the big problem is depository lenders just exiting some market areas, and leaving them to the subprime sharks. Suburban borrowers can find a HMDA-reporting depository or big mortgage company lender office on just about any block. Urban and rural and low-income areas generally find nothing but mortgage brokers. With brokers doing the "pre-screening" and the pricing, the AUS is just ratifying what the brokers do.
I do agree that raw race-based discrimination in loan approvals are much less of a problem than they used to be. But I do not agree that pricing is a level playing field across racial and economic levels.
You can automate underwriters, but until you automate loan officers and brokers, somebody with not just biases but a financial incentive in wringing the most out of the least sophisticated is mediating the decision.
Let me say something about screening, just to illustrate how actually very good intentions can harm certain borrowers.
Back in the days when I worked for a retail lender, we would charge a non-refundable application fee. That kept the non-serious folks out of the pipeline, and actually kept costs down to everyone (since we were compensated for the time and energy it took to do the analysis needed to legitimately turn down bad credit risks.) The application fee was equal to the cost of the credit report and appraisal.
Well, that's a fair amount of money for a low-income person to blow if they have no hope of qualifying for a loan. A lot of us used to go after the loan officers like the wrath of God because they were just extracting fees and wasting the time and hopes of people who should have been sent to a homebuyer counseling program and told to wait.
Well, that backfired for real, because it was like giving the LOs a license to screen people, and as Shnaps indicated above, when times were fat and there were plenty of doctor loans to go around, the LOs didn't want to screw with little bitty FHA loans to working people.
Nowadays, of course, "competitive" pressures have eliminated the upfront nonrefundable application fee. The result of which is that if you are misguided enough to apply for a loan, no one will let you get away without one because you must close for them to get compensation. You'll get put in some evil toxic loan if that's what it takes to qualify you.
So these things are not simple, and they have a lot to do with business structures and practices that don't have an obvious connection to race or class (that's the "disparate impact" school of thought, that considers how business practices that are not intentionally discriminatory can have disparate impacts on different kinds of borrowers).
1) "Magisterial" is the word that comes to mind to describe Tanta today.
2) Wow, an actual falsifiable hypothesis? Karl Popper would be very proud of Tanta!!
3) I am sure I have missed a thread or two, but speaking hypotheses, has FICO as a measure of outcome (servicing mortgage debt) been rigorously tested? To what extent is FICO a proprietary black box? And has there been grade inflation (has the method changed, is it grading on a curve?) Finally, has the predictive value of a particular score changed ovewr time? IOW, does a 680 from 2001 show the same behavior as a 680 from 2006 or is it too early to tell?
Tanta, thanks for bringing up some interesting considerations.
Sometimes I get so exasperated trying to explain that an AUS is egalitarian by design, I don't step back and ponder other systemic factors that may be surrounding the AUS.
I'm all for this blog as required reading. In fact, the good Dr. Carrington should invite you out for a guest lecture! With a paper bag over your head to maintain anonymity, of course.
Golly, David, you really want to go for the extra credit, don't you?
I shall try to write something on recent indications that FICOs aren't as well calibrated with mortgages as we have all been told.
For now, let me just say that there was, originally, a serious effort to do that. The GSEs got interested in FICOs in the mid-90s, and they started by just requiring them to be reported on loan deliveries. You didn't use them to make cutoffs or price loans or anything, you just collected the data and passed it on.
That allowed the GSEs, who have these big honkin' geographically diverse databases, to do some "calibrating" of FICO with human underwriting, in terms of qualification, and FICO with subsequent servicing history, in terms of performance.
The first result of that research, actually, came in GSE recommendations that FICO be used to set level of review for loan files. You didn't make the decision based on the FICO, but lower FICO loans had to be subjected to more detailed review than higher FICO loans. As an example, if the FICO was 720 or higher, all you had to look at in detail was recent (last 12-24 months) of credit history to make sure it wasn't trending down. If the FICO was less than 660, you really needed to wade through the whole picture, and get written explanations & more documentation from the borrower to account for the problems.
It was not until years after that that the GSEs started using FICO cutoffs for certain kinds of eligiblity. But they have never used FICOs in their own underwriting. Their AUSs do their own analysis of the full credit report file. People don't realize this: the AUS gets an electronic file of the whole credit report, not just a 3-digit FICO code. But when loans are manually underwritten, the agencies require certain baseline FICOs before you can offer certain kinds of maximum financing or other risk factors.
Eventually it gets hard to "recalibrate" FICOs to mortgage performance, because they are used to sort borrowers into various products (prime, alt-a, subprime), and so then you get FICO bands overlaying product bands, and then you can't tell which is driving what (the FICO or the loan terms themselves). It takes a HUGE database of loans to get significant results.
HUGE database. We're lookin' at you, rating agencies. If you expect individual lenders to do this analysis, it will be skewed by the individual lender's product line or geographical base.
Heh. I see.
Let's hope some congressional staffers are reading this.
The basic problem with sub-prime was that the lenders developed a bias based on "collateral dependency" which regulators consider to be unsound lending practices.
The belief and bias was so strong that "you can always sell or refinance". We see where that led to.
Yo Tanta. Much better. Our risk now is going to far the other way. The market is correcting itself. Lenders now have fear. But too much fear is a bad thing as is too much greed.
It all comes down to greed and fear.
The over pricing loans is increasing because many people are applying for loans. So the owner was increasing the interest so that he/she could have bigger profit. And the person who are engaged with this matter are very rich with income.
Thanks anyway...