First.
But all that means is I have nothing better to do that sit around waiting for CR or Tanta to put something up.

I'm an IT consultant, and I've always used the term "cheat sheet" as the one page of important stuff that you were allowed to bring into certain university exams, and by extension the one-pager to summarize whatever operation you're doing. "Cheat" doesn't have a negative connotation here.

Thirty years down the road and these things are as easy to fool as Barbie's My First Laptop?

To paraphrase Upton Sinclair, "It is difficult to modify an application to understand something when its purpose depends on its not understanding it."

Jail time and hard labor for the lot of them - execs included.

We. Are. Being. Robbed.

"Cheat" doesn't have a negative connotation here.

I actually suggested that it is ambiguous. Which I thought was a marginally more sophisticated thought than saying it is "positive" or "negative" in terms of "connotation."

But then not everybody took the same set of university exams, I suspect.

My question is: would you have used a "cheat sheet" in an open book take-home exam? Would it be OK for the rest of us to call you an idiot if you did?

A lot of industry participants are using "cheat sheets" in an environment that is in no way comparable to a timed examination. I think this is, um, curious.

would you have used a "cheat sheet" in an open book take-home exam?

why not? It is a lot faster to find information on a single sheet than in a 900 page book.

Tanta, admit it, maybe you just don't like anything that doesn't have many many words in it Smile

I, uh, can't get too detailed here-- but decision-making software is often subverted. Like, you're just learning about the system so you say to the guy in the next cubicle, "But I don't know the DIZF (Digital Intercommunal Zoophagy Factor)". Guy at the next cubicle says "It's always '6'". Or "How badly do you need to get to the next step?", or, ...

i work in a related industry (tech support for automotive insurance) and 'cheat sheets' or 'faq' or 'quick guides' are in use all over the place, mostly because the actual documentation with screen shots runs past 80 pages on occasion.

'cheat sheets' are a usefull tool even in untimed enviornments when they are correctly setup with a quick listing of items and catch phrases

for example 'repairable vs un-repairable workflow' is a good one. you CAN dig through the giant binder on how to use our stuff, or you can look at the quick start guide which tells you the one change you need to make to your workflow (check the box that says 'total loss')

I have cheatsheets all over my cube, but that's the only thing i'll defend.

the rest looks like I could have done better my sophmore year of college.
(if number_of_submissions > 1 and time_between_submissions < 10 minutes then Raise_red_flags_of_doom = 1)

why not? It is a lot faster to find information on a single sheet than in a 900 page book.

There are so many reasons why I never went into teaching.

The only "information" you can quickly find on a single sheet is the kind of information that can FIT on a single sheet.

I seem to recall instructors giving open-book take-home exams when they wanted something more elaborate as the answer than "12."

This was to my point about taking applications for the single biggest debt most people will ever incur being not really comparable, as a situation, to a timed exam.

My question is: would you have used a "cheat sheet" in an open book take-home exam? Would it be OK for the rest of us to call you an idiot if you did?

Guess I would be an idiot for using it on a take-home exam. But it has been useful to write up cheat sheets on, say, how to populate a new entry in a data entry app. The information may be all in the manuals, but it's a pain to look up all the time, or it doesn't tell you the best way to do things. Call it a "howto" if you want.

I don't care that it's called a "cheat sheet". I do care that it's a cheat sheet on how to to fraudulently stuff an application through the approval process.

I will say that my wife had 8 page 'cheat sheets' for the open book exams in college. It was simpler to have all the equations one might need ont he reference sheets than look through the source material and re-derive all the equations.

same thing with my programming. i have the stupid little plasitic tri-fold sheets with lists of reserved key-words and the like as well as some hand written ones about quirks of the compiler du jur's implimentation of pointer reference to sub-objects.

in short anytime you are given a program and told 'make it do x' there's going to be a knowledge base built up about how to massage it.

Matt could prolly get more into it, but it's often trivial to figure out how to really muck it up.

i work in a related industry (tech support for automotive insurance)

But "tech support" and "mortgage loan underwriting" are in fact very different things.

Of course I use this small-beer sense of "cheat sheet" to remember that the HMDA code for construction/perm loans is 7.

I'm losing my mind over the idea that "originating a mortgage loan" is mostly about such trivia.

Dear CR

"You think you will ever control for unethical behavior when you don't even demand moderate amounts of effort?"

You have summed up the entire Bush administration in one sentence. The mortgage industry took its cues from the top. So did Wall Street. And this is what you get, sooner or later. 'Nuff said.

Tanta

I have done process re-engineering for many years.
Once you agree to the process, you write the rules of how they need to operate. Some call them business requirements and related process flows.

As the propeller heads are cranky code,you are writting senarios to test the rules, and to test all the deviations (go back to your rant about IT a couple days ago).

After testing, but before you implement the change, you go through the documented policies and make the corrections. You also do change mgt to ensure everyone knows of the change happened and the impact to them.

Last but not not least, you have a set of tests added to the compliance checklist to ensure they are being followed. They explain what the test is, how to test it, and the required results. These audit tests have become critical in these days of SOX compliance.

That list of audit checks seems not to be the case here.

  1. if audit had been done with frequency, someone should have noticed everything looked too clean compared to histroical results. Smart mgt askes why nobody is reporting bonuses anymore and gets to the bottom of it.
  2. If they have this AUS system, then what is the problem data mining to tell shareholders the depth of the problem? Any decent analyst could go through that database looking for markers that flunked the rules and then give you an estimate of exposure. We have been waiting 9+ months for these crooks to come clean. They have been dodging and delaying while praying prices come back and this nightmare goes away.

Time to throw bankers on the bonfire with used car salesmen, professional ballplayers, politicans, mortgage brokers, and pediphile clergy as the honest ones do not want to express the outrage of some soiling their profession.

Who exactly is one suppose to trust anymore?
I am down to engineers and parents.

What is wrong with this country?

The first thing I thought of when CR posted that document was how a system like that could allow for multiple submissions of slightly tweaked applicant data without at least some kind of tracking system? (There's lots of reasons to track this kind of activity besides revealing "fraudulent-like" behavior. UI usability is one that comes to mind.)

Could it be a liability issue, Chase's version of "don't ask, don't tell?" My feeling is there was a conscious decision to leave out or disable the tracking functionality. These people can't be that stupid, can they?

Ah Tanta,

You would require people to think - read for understanding, internalize the content, and apply critical reasoning skills (by far the weakest link I am afraid) - unfortunately an activity assiduously avoided by the vast majority...

Tanta haz mad critical reasoning skillz!

I don't care that it's called a "cheat sheet". I do care that it's a cheat sheet on how to to fraudulently stuff an application through the approval process.

Well, I probably didn't make my point clearly enough, but it would have taken too many words to do that.

I am concerned about how the reliance on so-called "innocent" cheat-sheets seems to be connected to the phenomenon of "not innocent" cheat-sheets.

The fact is that I and everyone else have a whole collection of "innocent" cheat-sheets for using an AUS.

I have come to the conclusion that dumbing down the process to make it "easy" even for a newbie loan originator with no particular skills or training is part of an "enabling" process that allows certain kinds of fraud to thrive.

For one thing, since these noobs don't know why they're supposed to break out income, it's more likely that they can read this kind of thing and have it not immediately strike them as fraudulent.

When you expect very little thinking from people, you often get it.

Doesn't Zippy ever ask for supporting documents such as pay stubs, past year federal tax forms, financial statements showing the origin of downpayment funds? Zippy should require these things be scanned and added to the application. Zippy needs evidence or he/she is being a boob.

I have come to the conclusion that dumbing down the process to make it "easy" even for a newbie loan originator with no particular skills or training is part of an "enabling" process that allows certain kinds of fraud to thrive.

I heartily agree with this. Most innocuous cheat sheets assume the user understands what he/she is doing; it just summarizes how to use the complex, nonintuitive, quirky, and/or buggy interface to do what you want it to do. If you're telling the user to blindly type in a bunch of crap he/she doesn't understand, hell, I can write an autopopulate script to do that a lot cheaper.

As an example, a 747 has books and books of cheat sheets. "How To Fly an Airplane" shouldn't be one of them.

Could it be a liability issue, Chase's version of "don't ask, don't tell?"

Quite honestly, I'm ready to believe that it is directly connected to the bad training/no thought required/dumbed down universe they work in.

That is: in such an environment real error is enormous. They probably programmed some re-run limitations, only to discover that they were preventing "clean ups" from happening. The story would have been that if you prevent resubs, there's no way for some poor innocent broker to fix his errors.

Nobody asked, "Wy are there so damned many errors?"

Nobody said, "Well, tough noogies, folks. The price is too high. You'll just have to work harder on using the preview button."

Solution? Force the lender keep the loan on their books for 5 years.

I love you Tanta....

On a side note the link to the Oregonian is gone... Anyone have an alternate link?

Doesn't Zippy ever ask for supporting documents such as pay stubs, past year federal tax forms, financial statements showing the origin of downpayment funds?

No. No no no.

"Zippy" is the machine that decides who has to document income and who doesn't.

The whole idea is that it "identifies" loans that are "high quality enough" that the borrower can skip the coughing up of paystubs.

To require every loan to have documentation would mostly mean you don't need no Zippy. Because, you know, things no longer "zip" along.

I don't see that length is specifically the major issue. Good FAQs or cheats are hard to write a la "I didn't have time to make this shorter" adage. (I've certainly pulled out more hair over certain Abstracts than the entire paper they head.) They should hit the highlights, provide terse reminders of what you need to do in Frequent but probably not Everyday circs. The Full Docs are what you need to plow through to get the minutia of how to deal with a self-employed foreign national with haliotosis applying for a two-flat condo with a down-payment provided by his aunt's in-law's cousin. Or if you're new to the game and can't distinguish a garden variety request from a lunatic's diatribe.

The issue, at least to me, seems to be that that FAQ is a FAF, Frequently Applied Fraud doc. That and no one seemed to care to require whatever intelligence or respect for regulations or, ahem, common sense that supposedly once reigned in the business world.

T - oh sure, go and ruin the good fun I was having about this over at HW.

That said, you've got to admit: 1) the AE that thought this was tantamount to training is pretty dense; and 2) AUS platforms were never interested in high-cost fraud detection. We all knew that. Churn and burn, baby.

(Now, I suppose, it's mostly burn. But that's a different point.)

And we also know that lenders are finally realizing that spending money on fraud technology up front might actaully be a good idea. Shocking, I know.

I take from your post the rightful question of why in the hell did this take so long? And I couldn't agree more.

The only "information" you can quickly find on a single sheet is the kind of information that can FIT on a single sheet.
I once worked for a BIGLAW partner who refused to read anything longer than a page and a half. My question always was - when does it turn from editing into malpractice.

I have cheatsheets all over my cube, but that's the only thing i'll defend.

People who work in cubes need cheatsheets.

Cubes are too small for big binders full of crap.

When you have to spend most of your day in a cube, you do what you gotta do.

I don't see that length is specifically the major issue. Good FAQs or cheats are hard to write a la "I didn't have time to make this shorter" adage.

A lot of the time, cheat sheets are essentially different than the documentation. The documentation, especially bad stuff, says "this button does this, this button does that." It doesn't tell you HOW to do a particular task. Or it doesn't tell you that there is a specific workflow that you need to follow to make sure that nothing gets screwed up. For many apps, the training guide is a lot better than the actual documentation because of this.

Hair-pulling experience with a certain IBM suite of products deleted for space and sanity.

The story would have been that if you prevent resubs, there's no way for some poor innocent broker to fix his errors.

For this to be true, there had to have been a check and balance beyond a mere software flag: there was an auditing group and some kind of review that took place when a broker's error rate went above a certain threshold. Rather than disband the auditing group, they disabled the software that flagged questionable behavior.

The patients running the asylum. Smile

You would think they could take a more nuanced approach: with true data-entry problems, the errors would be mostly random and discounted. When the changes on resubmit happen in only one field, like Income, then it's time for a talk with your supervisor.

The Full Docs are what you need to plow through to get the minutia of how to deal with a self-employed foreign national with haliotosis applying for a two-flat condo with a down-payment provided by his aunt's in-law's cousin.

You should probably know that you are dealing with the person who used to sit quietly at her desk minding her own business when some random person would barge in and demand to know instantly how to handle a self-employed foreign national with haliotosis applying for a two-flat condo with a down-payment provided by his aunt's in-law's cousin.

And this person, moi, would look up, say "get a title policy endorsement over the condo legal description, I need an embassy or consul seal on the borrower's affidavit, and put two month's worth of bank statements from the aunt and the cousin in the file before underwriting gets it. Plus three extra points, another buck-and-a-half on the rate, and don't come back in here unless you've discovered where they hide the free mints in plain sight down on the teller line."

Yes, I am pissy about the number of years my good nature was taken advantage of.

Nonetheless, I'm living proof that after a number of years of exposure to your own business you can learn even the hard rules.

It's OK for people to fail to meet very high standards. It's inexcusable not to have high standards.

Thanks, Tanta. I love your work.

Unfortunately, I now think Zippy is a total sap.

It's all about short term incentives. Remember, back in 2005, everyone agreed that real estate only went up and appreciation would cover for our mistakes. You could be virtuous, rational and professional, and do 10 deals a month, or you could wink at the INDUSTRY STANDARD LEVEL OF FRAUD that was a fact of life back then, join in the race to the bottom and make a pretty good living. Since you were selling (and insuring) this toxic crap, what did it matter.

All the incentives were in one direction.

You would think they could take a more nuanced approach: with true data-entry problems, the errors would be mostly random and discounted. When the changes on resubmit happen in only one field, like Income, then it's time for a talk with your supervisor.

I bet there's a cheat sheet on how to change your approach to this from time to time so that you don't establish the pattern.

Like, sometimes start with very very high income and work it down. This way they won't zero in on your "inch up" strategy.

Well, I think this definitely started as a joke. After all, it's now clear that option ARMs, CDOs squared, SIVs, and most NAR press releases had their beginnings as mere jokes. It's clearly the fault of whoever took the good fun seriously!

Look, the memo simply is (if real) a Chase memo describing how to defraud Chase. Which ought result in Those Responsible Being Sacked, though that probably happened a while back owing to the general lack of need for their services.

It doesn't really read like a joke, having seen many (and wrote a few) back in the day of running analytics and pricing for an MI, it's not near scatalogical enough for that genre.

But the AUS developed from a risk management engine to a secondary market sorting system. It really didn't matter whether, y'know, actually lending this money to that guy on these terms made any sense, just whether or not that collection of facts amounted to a salable loan. So I cant be surprised that they were less than vigilant in enforcing its integrity.

As someone who random people cube-visit to demand the difference between Municipal and New England Townships and which do they have in California? and why exactly can't we make a ZIP-9 fold map? I empathize.

off topic

Troubled bank and mortgage lender Fremont General said on Friday that Federal regulators had stepped in and given it 60 days to raise additional capital, or face a forced sale. The Federal Deposit Insurance Corporation issued a so-called Supervisory Prompt Corrective Action Directive to Fremont General and to Fremont General Credit Corporation, the company said, requiring the bank to take immediate steps to recapitalize.
In March of last year, the FDIC issued a Cease and Desist Order to Fremont regarding its subprime lending operation, including an allegation that the company violated Section 23B of the Federal Reserve Act by engaging in transactions with its affiliates on terms and under circumstances that in good faith would not be offered to, or would not apply to, nonaffiliated companies.
Under the terms of the new directive by its regulators, the bank has 60 days to either raise enough equity capital or otherwise sell obligations in order to reach FDIC

off topic

First shot across many a bow.

Well, I think this definitely started as a joke.

Let me be clear that I meant "back office black humor," of which there is often surplus supply.

I have myself written a few wicked, wicked memos. I had two purposes besides wasting time and indulging myself:

  1. To provide some much-needed comic relief to my trusted colleagues, and
  2. To see how many of the idiots fell for it.

These days most of us get that kind of thing out of our system by blogging. And we still find that some people enjoy the satire and some people fall for it, every time.

In any event, I was not attempting to "minimze" the thing. Nor does Jonathan Swift really believe in eating Irish babies.

Tanta,for the last few years anyone who insisted on high standards did not do any business,bad money still drives out good.And as far as the drones in the back office,as you have noted in the past,the old timers cost too much and were frequently forced out and replaced by temps or part timers with no background or training.Even in the old days people who insisted on doing a good job and being prudent were tolerated at best...

Investors fear impact of Fitch CDO rating overhaul
| Reuters

see if fitch goes ahead and revises ratings on synthetic CDOs. many on wall street are a-feared. the tinker-bell game goes on (close your eyes and just believe)...for now... but the monkey on the addicts back just grows stronger.

And as far as the drones in the back office,as you have noted in the past,the old timers cost too much and were frequently forced out and replaced by temps or part timers with no background or training.Even in the old days people who insisted on doing a good job and being prudent were tolerated at best...

Tom Stone | 03.28.08 - 11:52 am | #


Sounds like the FDA, CDC, NASA, FEMA, NSA, CIA, FBI, GSA, The Pentagon, and all of the agencies involved in the regulation of banking.

Now where would big business get an idea to do something like this?

It's a pity that at some point in one of my moves I lost my only copy of a masterpiece I wrote back in 1999, or I'd share it with you all.

It was a set of specs for Mortgage Asset Due Diligence Operating Guidelines, or MADDOG 2000.

I bet it wouldn't sound quite as ludicrous today as it did back then. In fact, I bet I could get research funding for it today.

Excellent work, Tanta, as usual. Between the human phrasing and your subsequent commentary, you never fail to personalize what could be really thick going. Your backroom adventures are never dull, and never fail to bring a rueful smile to my face. Makes burning out the bearings on yet another scroll wheel a pleasure.

On the subject at hand: You seem to be going out of your way to remain non-judgmental. I suppose given the dearth of details or the provenance of memo, you have little choice.

So, to further cloud the supposed motives of the memo, is it possible that this represents the "actual working policy" - the steps needed to get the damn loan past the AUS - which is the necessary "official policy" needed for the inspectors and regulators?

(Viewing from the outside,) there does seem to have been a dogged determination to get very damn loan possible approved during the maelstrom of qualifying mirror-foggers.

While I've never seen this written down, I've seen efforts like this to get by both human and automated underwriters.

The efforts were system wide. Not just Chase.

I write a little about it in the link above.

From wikipedia: AUS or Aus can refer to the following items:

* Australia, the ISO 3-letter country code
* American University of Sharjah
* Application Update Service, the service for updating Mozilla products
* Lauri Aus (1970-2003), Estonian cyclist
* Association of University Staff of New Zealand
* Austria, a nation in central Europe.
* Austin-Bergstrom International Airport in Austin, Texas
* Atlantic University Sport, the governing body for university sport in Canada's Atlantic provinces
* A village in Namibia about 120km east of Luderitz, with an approximate population of 200.
* Banu Aus - One of the Arabian tribes that interacted with Muhammad
* A preposition in the German language.
* An abbreviation for AUS, Army of the United States, the wartime conscription component of the United States Army
* &#256;.U.S.-&#256;nkh, Utcha, Senb for the Ancient Egypt phrase: "Life, Prosperity, Health",–(Ankh, Utcha, Senb).</i>

Based on what I've read in this thread, I think the most suitable spot for a notation on underwriting would be between the Estonian cyclist and Kiwi college staffers.

You see, the more things change, the more they stay the same:

"We have come to be one of the worst ruled, one of the most completely controlled and dominated, governments in the civilized world - no longer a government by free opinion,no longer a government by conviction and the vote of the majority, but a government by the opinion and the duress of small groups of dominant men."

Thomas Jefferson,1816

Where, in other words, is the "high" tech? We've had AUS since the green-screen mainframer days of the 80s, kids. Thirty years down the road and these things are as easy to fool as Barbie's My First Laptop?

Tanta -- of all the things you've written, more than any I loved the one a few weeks back where you let loose on the mortgage industry mgmt for cost cutting, and 'efficiency', and 'outsourcing', and how just maybe all that was coming back to bite them.

Well... guess what? the managers at lending institutions went to the same schools that the managers at software companies did -- and learned the same lessons: cut costs, make short terms profits, collect big bonuses, get promoted or move to another company, bail out before the problems hit. The software industry is at least as rife with the kinds of superficial efficiency, and complete lack of interest in long term consequences, or indeed responsibility as any industry in this country. Indeed, unlike most any other industry (except nuclear power intruigingly enough) they are legally not liable for any consequences arising from the use of their product -- take that mortgage people! As you no doubt know, ever since the 'liberal' 90's we've been shipping tech jobs, software jobs, to india, after all... they are cheaper. And management, I can quite assure you, pats itself on the back for showing all those stupid geeks that all their mealy-mouth concerns about reliability, and safety, and useability, and customer support, and customer loyalty and branding was just a bunch of stupid geek talk that didn't do nothing for the bottom line. I'm actually in India right now (visiting not working) and I can tell you one thing -- there is absolutely no way that most of the software people trained here are anywhere near as good as americans -- the schools just don't have the resources to provide the same level of care, information, knowledge, nor frankly is there anything like the competitiveness, or numbers of students. Sure, some indians are extremely bright, and yes, some are extremely good at technical things. But destroying americas technical class to make a short term buck that only benefits the managers is another crisis, that while hidden, and while it will never be as plain as the housing one, indeed may never really be put together conceptually as a crisis, much less where it came from, will over the long haul, prove even more devastating to the american culture, and american economy, and american society, as the nightmare in housing created by short sighted, excessively empowered, excessively freed of responsibility, hardy har-har, aint' life grand, good ol boy, MBA's.

On the subject at hand: You seem to be going out of your way to remain non-judgmental. I suppose given the dearth of details or the provenance of memo, you have little choice.

Certainly, I dislike drawing firm conclusions ahead of the facts. It's not like we don't live in Innuendo World these days where people become convinced they "know" things that they only kinda suspect because some reporter played free-association. So in that sense, yes, I think one should be cautious.

But it's mostly that I'm actually more worried about it being a "joke" that a real live AE didn't "get."

This gets back to my long unsuccessful war on the idea that conscious cynical malevolence is the only explanation for what we see, not badly organized befuddled incompetence.

My usual assumption is that if this really were intentional policy on Chase's part, they'll get busted (they'll surely get sued from here to eternity by bondholders) and that'll be that.

But if it's a case of a joke policy being so close to real policy that Chase AEs can be confused about it? They'll get a pass for that.

I'm trying to show why they shouldn't get a pass for that.

You are my hero, Tanta! You've got my vote for head of OFHEO, OTS, the Fed and the FDIC, all at once!

Great post, Tanta.

Another thing I think that needs to enter into this. Major institutions tend to have checks-and-balances in place (hence the tendency of smaller shops to say, "Well, if Chase is doing it this way, it must be ok.") Their senior people also usually have the sense not to commit to paper anything that has any tendency toward autogluteal mastication.

So, when one finds a stinker like this memo coming from a major institution-- even as a joke or half-joke-- one can guess that practices at smaller places are much, much, much worse.

Tanata - I see your MADDOG and raise you the Automated Scoring System for Home Ownership Loan Excellence (circa 1995). "Does the borrower qualify?" "Well, I'll just have to look at his..."

There are some very interesting questions here. I have a little 'cheat sheet' right here at my desk - written up by me - that is a guide to building code requirements for accessibility. It even says 'cheat sheet' right at the top. But it cheats on nothing; it is just a quick way to verify that you comply, not a guide to how to avoid compliance. But the cheat sheet is helpful because the actual code information - in this day of computers and quick access - is difficult, arcane, technical, hard to follow and generally annoying to use. Why should that be so? There are several answers - one is that jargon, in any field, is SUPPOSED to be difficult because is defines and justifies the 'priesthood' - the special people who, alone, are the interpreters. It is their job security.

Voicefromthewilderness,

A contractor came to talk to us about our kitchen renovation. He said in any project, there are three things people want.

  1. Low Cost
  2. Good Quality
  3. Fast completion

He said, I can give you two out of the three.

(There is of course a competence assumption.)

The same is true of software, regardless of where it's developed.

Regards,
Pat

So, when one finds a stinker like this memo coming from a major institution-- even as a joke or half-joke-- one can guess that practices at smaller places are much, much, much worse.

In my experience, the major institutions actually hire professionals--sometimes like me--to write stuff with minimal autogluteal mastication potential (I'll be stealing that one from you for the rest of my life). And their stuff is pretty good, on the whole.

The smaller outfits plagiarize it, because they can't/won't afford their own literate professionals.

They often plagiarize it badly. You can actually hold the documents up to the light--one from Chase, one from Loans R Us--and see where the latter left out a few key words (like "not" and "never" and "ineligible" and so on).

Then some real dipshit takes the bad plagiarism and turns it into a "cheat sheet."

Then the "cheat sheet" makes the e-mail rounds, and some idiotic broker sends it to his Chase AE with the suggestion that it'd be a great "training tool." The Chase training tool--I mean, AE--is dumb enough to recirculate it.

So it goes full circle, and suddenly Chase is the "owner" of a fourth-generation corruption of Chase policy.

I used to get AEs sending me the most patently idiotic stuff with suggestions that I use it for "training materials." It wasn't jokey, and it certainly wasn't recommending fraud, it was just wrong because it was oversimplified or out of date or based on a misunderstanding of a term. (But hey! it was easier to read than the stuff I put out!)

I'd be inclined to think it would take a fairly long trail of breadcrumbs to get to the "source" of this one.

This gets back to my long unsuccessful war on the idea that conscious cynical malevolence is the only explanation for what we see, not badly organized befuddled incompetence.

The former was my intent - perhaps what started out as word-of-mouth-get-it-done guidance by a non-Tanta type manager, turned into the memo by the latter - "badly organized befuddled incompetence" would be quite capable of not remembering verbal instructions without writing them down. Once something is written down, the rest is history.

But if it's a case of a joke policy being so close to real policy that Chase AEs can be confused about it? They'll get a pass for that.

You're more generous that I am. But then incompetence in my racket costs lives and/or limbs.

... oh yeah... The mainframe's still alive and kicking. The green screen's been replaced with a GUI or browser for everyone but the administrators (and some developers).
(Pat clances fondly at the green glow emitting from monitor #1.)

My father is a retired engineer. I remember when that overpass collapsed in Minneapolis, and I asked him what he thought about it. He said before computers, engineers had a tendency towards overkill. When trying to determine load tolerances and other calculations they always rounded up and tended to make things stronger than necessary. Once computer drafting started taking over in the 70's engineers were able to make much more precise calculations and thus the emphasis shifted to building things with as tight tolerances as possible in order to minimize the cost of getting the job done just to the standards that were deemed necessary. But in the case of this overpass, since the time when that bridge was designed and built, the length and weight limits on tractor trailers has been raised, there are now dual-trailer rigs and heavier traffic. The design of the bridge to deal with stresses that were not intended and over time overstressed the bridge causing catastrophic failure.

Analyzing risk is the same, when things can't be calculated precisely smart one tends to err on the side of caution. Analysts and executives became overly confident in the ability of their computers and models. This overconfidence cut into that margin of error, and now they realize "Our models didn't account for this or that possibility." Models and computers are a tool. They are still very much subject to human error, but people tend to forget that and rely on them as if they were infallible.

One thing I never got about the gift rule was how the money magically became clean after two months. I had a small amount of gift money in my mortgage, less than 10%, but my lender nearly turned white at the mention of it. Then it turned out I had been holding it in a bank account for six months. So my six month old gift wasn't a gift, it had magically converted into savings after the first two months... and savings were ok.

I'll accept money not earned means less to some people. But a six month old gift wouldn't seem to mean more than one received today... at least not to me.

guest,
Another consideration is that the models are always based on history. In engineering, that can be a stable, defined thing. In politics and economics there can be wide variation... and we are seeing very poor performance lately by all sorts of models - the government being as lousy in tracking things like jobs and cost of living as the 'risk' projection models were.

Alex

Um, is your comment supposed to relate to what I wrote? 'cause I sure don't get it.

I have told managers that exact same thing, and had them look at me with that hard nosed, 'tough guy' manager look at tell me they want it done yesterday, at half price

But you are absolutely right, they love ignoring quality. which, um, was my point.

Quite frankly, all you had to do was start out your post with the word 'contractor', to tell me all I needed to know about how much insight to expect. And attention to detail, but as too cost, you don't really think that contractors are 'cheap' do you? I got bad news for you -- if you keep getting your 'knowledge' from 'things contractors told you', you might be in for a surprise.

One thing I never got about the gift rule was how the money magically became clean after two months.

Actually, it's about making sure you didn't borrow it.

After two months, if you had borrowed it, it would have shown up on your credit report.

The trouble we have with people who get gifts right at the time they buy the house is that you sometimes can go back two months after closing, run a new credit report, and see that the "gift" was actually a loan. It was called a "gift" to either avoid counting the payment in the DTI, or avoid the CLTV limit (the limit on how much of the purchase price can be borrowed from all sources).

My analogy is agnostic to where the software developers are. The pressure for speed and cost over quality knows no national boundaries.

Although off-shoring something that requires cultural awareness to handle accurately could present challenges to the competence assumption.

Regards,
Pat

I don't get it.

If everyone follows these "rules", as you call them, then how are people going to buy houses?

My local CEO needs his bonus!

Alex

My point was that there is a social cost to redistruting labor overseas, and that the cost will not be paid by the people who made the decisions. It was something of an analogy with the mortgage situation and the banking situation. My other point was that software sucks for a reason, because management chooses short term costs as it's only metric for 'goodness' of software, and that there are systemic reasons for this.

I'm not sure if you are aware of how arrogantly you are expressing yourself for someone who completely missed my point.

But thanks for the lecture on how quality works, and engineering is done, direct from you kitchen contractor.

At the begining of the thread, I think there is a fundamental tension and misunderstanding of the term "cheat sheets".

There is one sort of "cheat sheet" which is essentially a home-made FAQ or FAActions. Then there is another sort of cheat sheets. Those provide premasticated answers.

Do try to keep those differences in mind before saying cheat sheets are fine! The English Language Is Frail.

Hey Tanta, In the Race to the Bottom, I remember seeing a deal where a 1950's style Motel (single story, 20 units) was "converted" to a Condo. Thought I had seen it all at that point until I saw a deal for a Condo-Hotel.......

Ahhh, Good Times............

...the competitive pressures that spawn "races to the bottom"

That's a great summary in itself.

It's OK for people to fail to meet very high standards. It's inexcusable not to have high standards.
Tanta | 03.28.08 - 11:23 am | #


Tanta:

A serious note.

Thank you for giving fairly simple words to something that we've been trying to get the kids to see.

Trust me, this will be disseminated.

In any event, I was not attempting to "minimze" the thing. Nor does Jonathan Swift really believe in eating Irish babies.
Tanta

I guess the that would be comparable if there had been a lot of Irish-baby eating in the early eighteenth century. On the other hand, if the Modest Proposal had instead been about imagined forced removals of Scottish tenant farmers, I could see the similarity. But given the outcome, I think the public of the time might be equally skeptical about the proposal being a mere joke.

There are so many reasons why I never went into teaching.

One of my very best instructors told us we could bring one page of hand written notes with us to the exams. Anything we thought would be helpful, hand written as small as we thought possible. Formulas, principles, examples, anything.

Her test questions required us to apply all the stuff we were supposed to know and could look up, either on our faq sheet or latter in life, the binders on the shelf (or google nowadays) to situations we had not been walked through in class but could reasonably expect to see in the professional world. The students who could/would apply critical thinking skills to the problems did well, the lazy ones who thought all the answers were on the faq sheet did poorly.

You are teaching here and we appreciate it.

What business hasn't got tensions between and among sales, accounting, compliance, senior management and, more recently, IT?

Surely some of us envisioned a lively future the first time a software writer asked us to describe our work in detail, i.e., "Tell me everything."

As an IT professional, on projects like the one you suggest, my first question is "Why are you hiring so many people who are willing to bend the policy rules so much to get iffy loans approved? You could save yourself $50 million just by not hiring assholes."

But I already know the answer - "We want strict policy, but we don't want it to get in the way of making money."

High-tech can never solve that problem. And the problem with high-tech is the sheer amount of work involved when policy changes. The fact is policy changes VERY quickly, often in unforseen ways; if you want a highly configurable-on-the-fly automated solution, you are talking huge expense.

Most of the reason China is kicking the US's ass in manufacturing now is precisely because China IS NOT automated to the hilt. When the assembly line needs to work on a new model, there's no shutting the factory down while the robots are re-programmed and re-tested; humans are MUCH more adaptable than machines.

And that's precisely how cheat sheets come into play. I can foresee all the behavioral logic being programmed into an AUS, and instead of a cheat sheet being generated by the worker bees in a matter of days, it'll take a matter of weeks. But the point to remember is the cheat sheet will STILL be generated and it won't take years.

Somehow all this "race to the bottom" stuff has me thinking of the sky diver whose parachute did not open. He is thinking, "well, I am going really fast here and no problems yet..."

Did senior management think that, just because they were getting some really good performance bonuses for skipping basic underwriting steps, life would continue "no problems yet..." forever?

Why wouldn't a known design flaw get fixed?

The same reason someone removes the safety equipment from industrial equipment. Someone took the safety guard off because it was slowing them down. It won't get put back on till someone get's their arm pulled off.

The phrase I would hear more often than I care to admit when trying to get answers to problems like this was "It's a business decision".

Tanta, sometimes I think you ARE Dave Barry.

Surely some of us envisioned a lively future the first time a software writer asked us to describe our work in detail, i.e., "Tell me everything."

I learned fairly early on to always take those questions very, very, literally.

Two hours later, after we've gotten past the point where I took off my coat and hung it up on the peg and powered up the PC, they have often gone away. I'm still out two hours, but I haven't had to explain what I do all day. Fair enough.

Thank you for giving fairly simple words to something that we've been trying to get the kids to see.

Trust me, this will be disseminated.
homedad43 | 03.28.08 - 1:20 pm | #

Good luck with that,...a hundred fifty years has still failed to reach the masses living in quiet desperation.

In the long run men hit only what they aim at. Therefore, though should fail immmediately, they had better aim at something high. Thoreau

The chapter on Economy, in Walden, has quite a few pithy references to our present situation for those who are ready for it.

Where, in other words, is the "high" tech? We've had AUS since the green-screen mainframer days of the 80s, kids. Thirty years down the road and these things are as easy to fool as Barbie's My First Laptop? After all the money these lenders have spent over the years on IT? There's something else that doesn't add up here besides a borrower's paystubs.

"IT" and "High Tech" isn't the answer to every problem and is probably completely to blame here, along with their management of course since failure is a team effort.

The systems you are describing and desire are HARD to conceptualize and on top of that require either a lot of experience in the subject domain or truly gifted "IT" people who can tell when they are asking the right questions to the wrong people and when they are asking the wrong questions to the right people.

Most IT projects fail for this reason chiefly. Even "successful" IT projects usually fail a business in the long run if you stop to calculate the cost vs the cost it saves.

Every single person here has been exposed to the IT project that is a negative productivity gain I'd wager, and will continue to be. You will continue to not get what you pay for, get used to it.

VoiceFromTheWilderness wrote: My point was that there is a social cost to redistruting labor overseas, and that the cost will not be paid by the people who made the decisions.

Actually, there's a social benefit. Adam Smith 1776. What's supposed to happen is that the better educated American is supposed to get a job providing more marginal value than his previous one, rather than mooning about "social" costs in the blogosphere. Should we be self-sufficient in banana production, too?

Russ wrote: As an IT professional, on projects like the one you suggest, my first question is "Why are you hiring so many people who are willing to bend the policy rules so much to get iffy loans approved? You could save yourself $50 million just by not hiring assholes."

Yeah, I'm gonna take that insight to Vegas. They could save 9 figures on tech, easily, if they just paid their blackjack dealers more.

The whole industry has some explaining to do.

When you have to explain a joke it's not funny any more.

Chase mortgage memo pushes 'Cheats & Tricks'

The bank says it never backed the strategies, which detail how to get an iffy loan approved

Story not found - OregonLive.com

JEFF MANNING
The Oregonian Staff

(Thursday, March 27, 2008) A newly surfaced memo from banking giant JPMorgan Chase provides a rare glimpse into the mentality that fueled the mortgage crisis.

The memo's title says it all: "Zippy Cheats & Tricks."

It is a primer on how to get risky mortgage loans approved by Zippy, Chase's in-house automated loan underwriting system. The secret to approval? Inflate the borrowers' income or otherwise falsify their loan application.

The document, a copy of which was obtained by The Oregonian, bears a Chase corporate logo. But it's unclear how widely it was circulated or used within Chase.

Bank spokesman Tom Kelly confirmed that the "Cheats & Tricks" memo was e-mailed from Chase but added that it does not reflect Chase corporate policy
.
"This is not how we do things," he said. "We continue to investigate" the memo, Kelly said. "That kind of document would neither be condoned or tolerated."

The March e-mail was sent by Tammy Lish, a former Chase account representative in Portland. Chase fired her days after discovering she had sent it.

"I did not write it," Lish said. "It was sent to me by another (Chase) rep in another office along with some other documents that were more step-by-step customer training documents."

Even if the memo was penned by a single employee, it illustrates an attitude prevalent in certain corners of the mortgage industry during the boom years. In the face of sustained and significant home price increases, much of the industry veered away from traditional notions of safe and sound lending. Loan volume became as important as loan quality, particularly for the rank and file typically paid on commission.

During the boom, it was common for lenders and brokers to get paid more for risky subprime loans than for 30-year fixed-rate loans because the higher-interest loans fetched a higher price on Wall Street…”


VoiceFromTheWilderness writes:
My other point was that software sucks for a reason, because management chooses short term costs as it's only metric for 'goodness' of software, and that there are systemic reasons for this.

Stop pushing the failure of IT projects up the chain.

When you take your car the the mechanic and he doesn't fix it right the first time do you blame yourself for wanting it to be done efficiently and inexpensively or do you blame the mechanic for being bad at his job?

There reason we don't blame ourselves is a combination of ego and self preservation. Not many of us are confident enough we are better then the average to demand that salary premium to continue living the lifestyle we've grown accustomed to thanks to the IT boom of the last few decades.

Housing Wire said the memo was a joke. I couldn't tell if they were being sarcastic.

Here's what pissed off Chase Wholesale AE's about ZiPPY SISA:

1> They couldn't use it. They had access to a test system only. They couldn't run live deals through ZiPPY to figure out how to game it. Instead, they would have to borrower one of their broker's Login's and passwords to the real ZiPPY system.
2> ZiPPY SISA sucked compared to CountryWide's Fast and Easy. Fast and Easy was proactive, ZiPPY SISA, reactive. Try walking into a brokers office after the Countrywide Rep had just been by. The broker says they're sending everything to C'Wide because of Fast and Easy. THe Chase AE says they have the same thing: SISA. Great the broker says, can you get me the guidelines. AE: "Oh, no. It doesn't work that way. You just run the loan through and ZiPPY will let you know if you got a SISA. Hey, would it help if I gave you a Chase golf shirt?"
3> Pricing sucked. Of course, AE"s always complain pricing sucked, but the point here was that ZiPPY SISA, being reactive, theoretically, should have a given Chase a better commitment with Fannie/Freddie. That never seemed to show up in the pricing.

I love the fact that Chase is getting soiled. It was unfair that Countrywide was taking all the crap. After all, Chase Mortgage's motto was "We need to be like Countrywide."

So, to be more like Countrywide, enterprising AE's at Chase did the difficult work of borrowing passwords, inputting a bunch of crapping loans and figuring out how to get one over on ZiPPY. They passed their thoughts past friendly underwriters--many of whom didn't work for Chase, but were contract underwriters. Then the AE's went out a wrote their on ZiPPY Cheats and Tricks, also known as Chase Fast and Easy.

Both before and after automation, lenders asked for incredibly stupid things.

Closers in the trenches like me, would have to figure out how to actually close a loan, given impossible conditions in many cases. And requirements that if you took them literally, would have put the ki-bosh on half the loans. I've been doing this a long time and so know the difference between a work-around and a mortal sin and a venial sin; things you can safely overlook and things which will have horrible results. The thing is, while giving loans they shouldn't have the lenders were also making ridiculous requirements, that made no sense in terms of the ability to pay back a loan.

The second stupidest closing requirement was when I had to have the borrowers explain why the didn't pay a three dollar (yep $3.00) medical bill. They looked at me blankly. I made up a fairy tale and they signed. It is my considered judgment that a $3.00 unpaid bill is completely and totally irrelvant to the ability to pay back a loan. I would do the same thing in a second, if I had any loans to close.

I actually have forgotten the stupidest thing, but it was considerably stupider than the $3.00 thing.

When you have inexperienced closers, then they get the idea that the fairy tale is ok in all cases. Same all up and down the line.

The reason I can't believe that this "cheat sheet" is for real is that it is so simple minded. Ignoring the honesty issues completely, who wouldn't have memorized those things in the first morning of work? Real cheat sheets show you how to do things that are slightly complicated and that you don't do every day so that you memorize it. How to import a tricky file into your document; not how to bold a word.

Fascinating! Having been on compliance staff at multiple major financial institutions over the past 10 years, it's fascinating to see that everyone, everywhere is dealing with the same basic issue: expediency triumphing over basic principles.

Why did Rome fall? Because those in power no longer cared about the principles that made Rome great.

Wheeeeeeeeee!

"along with some other documents that were more step-by-step customer training documents."

I assume that 'customer' in this context refers to brokers.

There is much truthiness in this post and comments. We know it gets archived, but I'm leaving breadcrumbs back to this article.

FWIW, I used to be a loan officer. I considered myself upstanding and tried to get everybody the best deal they could qualify-for and priced it for 1.5 points yield spread. This typically resulted in an interest rate equal to or better than retail, and I attended closings and made sure things went as best they could.

Sometimes, in a rising home price environment, getting an appraisal to match/support the sale price was difficult because prices were rising.

Once, fiddling with DO, we put in a higher home value by about 40k, and DO no longer wanted an appraisal. Since this was a refinance, and we were guessing, we took the findings and closed the loan.

Fraud?

Didn't feel like it. Messing with DO or DU to get an approval is what originators do. If I can't get you a 90LTV cashout refi, I'd try 87 or 85... something to see what works.

My 2 cents. That memo was not a joke.
Everybody gets paid at closing, so everybody wants to close the loan.

also, note to the authors, I appreciate your site and read it a lot these days b/c we are looking to buy a house this summer. Assuming there are mortgage lenders extending credit this summer. Smile

ed
Why did Rome fall? Because those in power no longer cared about the principles that made Rome great.

Amen. you bring the marshmallows I will bring the hot dogs. I feel the embers burning.

In any event, I was not attempting to "minimze" the thing. Nor does Jonathan Swift really believe in eating Irish babies.
Tanta

At least, not the skinny ones.

Fraud?

Didn't feel like it. Messing with DO or DU to get an approval is what originators do.

And that, my friends, is what I have spent eleventy jillion words trying to say.

If it doesn't "feel like" fraud to someone whose goal is only to get a loan closed, then it just isn't fraud. QED.

After all, there can't be any reason why the AUS came up with a different decision on the higher appraised value. Right?

And after all, nobody "meant to" be a sleazebag. All we "meant to" do was give a customer what he asked for in a painless fashion that ensured we got our back-end points. Fuck the lender or investor or rating agency or whatever sad sack pension plan these loans end up in. They ought to look out for themselves, right? If they let loan officers fool them, then who cares, right?

I'm about to lose my lunch.

I will say I can see why so many people might be resistant to the idea that this started out as satire. If you don't think the underlying conduct is a problem, you probably can't see why some wag in the back office might toss together a "Fraud for Dummies" training session "handout" to take a few digs at you.

Clinton said: "Messing with DO or DU to get an approval is what originators do."

I watched that happen ALL the time. In banks. Just because someone originates for a bank doesn't make the holier than the unwashed broker masses. Don't kid yourself.

"the holier' = 'them holier

Tanta, don't lose your lunch.

You are absolutely right in pointing out that too many people fail to see the ultimate consequences of their actions, and therefore believe those actions to be justified.

There's a natural ebb and flow in the course of events - sometimes, like now, the actions of the majority of short-sighted individuals result in chaos and destructions.

Those times of pain are precisely the time when people start to learn to look more deeply at what they do - and people like you who have been trying to teach us become ones who lead.

If you do lose your lunch, I'm sure there will be many of your readers who are willing to step in and help clean it up, following the example that you and CR have set.

Thanks for being an ounce or calm rationality in the midst of the madding hordes...

I never said fuck the pensioners. And I meant no such thing.

Maybe my point can be recast: DO/DU or LP or any other system --as used by wholesale brokers -- was (where I worked) an iterative system where inputs lead to approvals or refusals.

I am not saying who cares.
I am not saying that it was good. But that an iterative system allowed for guesses to be re-guessed.

Who owes a fiduciary duty to retirees and clients and such? I think the people rating the securities, or investing the pension money have the duty to look closer than an AU system that didn't look very hard. Whoever sanitized the crap and gave it a AAA rating based on home prices going up forever.

Again, I only chimed in to say that the memo doesn't look like a joke. All that stuff actually happened.

AU/DO let you guess. It was guess or wait 3 weeks. I suggest that many guessed. I don't suggest it's a great way to run a financial system, but either is 100LTV Stated NOO.

but yeah, okay, maybe the memo is a joke.
...
who's laughing?

Is the memo real? As the great Homer J. Simpson says often "it's funny cause it's true"

Well, I can understand writing a cheat sheet to get around a program written by a bunch of brain-damaged idiots who never thought of the possibilities of X. (Remind me to tell you the time the MIT Solar Car Team tried to get the Aztec registered in Massachusetts. "Number of cylinders"= 0 is definitely Not Acceptable to the computers down at the DMV.)

Tanta,

Best post ever! And, I've passed it on to all the underwriter's I know, and what Senior Management that will receive my e-mails! We've known since 1995, when Fannie and Freddie came out with DU and LP that there were people "gaming" the system, and despite our protests, Management did nothing. Then, all the Lender operated systems, like Zippy, Clues, etc. joined the party and again, the "gamers" came out to play. Meanwhile, back in the trenches, underwriters came to expect the Management override. You are right: "This industry has some explaining to do!"

Garbage in, Garbage out still seems to be holding true then.

Login or register to post comments