Saturday, February 6. 2010
In his remarks before the American Securitization Forum on February 2, 2010, Comptroller of the Currency John C. Dugan made the case that minimum underwriting standards could play a major role in reforming securitization markets, and, in fact, he went much further and made far greater assertions -- assertions that are strikingly similar to those expoused in the CCAF white paper, my Congressional Testimony on July 28, 2008, and the book on the new lending system that was published last April. Comptoller Dugan said: "I think there is a better and more direct way to improve underwriting standards, at least for residential mortgages – and to do so not just for the benefit of securitization markets, but for all mortgages whether held or sold. Instead of going at the underwriting problem indirectly through “skin-in-the-game” requirements, why not attack it directly? If quality underwriting is the goal, then why not, as I’ve suggested in a previous speech, establish minimum underwriting standards directly by regulation, at least for residential mortgages? Why not apply these standards to all mortgages, whether retained or securitized, so that there is an entirely level playing field? And in the context of securitizations, why not stipulate that if the standards were satisfied, there would be no need for skin-in-the-game requirements that could defeat true sale treatment of securitized assets?" What Comptroller is saying is that if the proper underwriting stanndards are enforced, then there is no worry about the "originate-to-distribute" model, i.e. if the loans qualification process is solid, then it doesn't matter whether the originator of the loan sells it or keeps the loans it makes. With good underwriting systems, lender skin-in-the-game is not needed as an incentive to make good loans. We can conclude that the problems ensountered with the "originate to distribute" (OTD) model were not due to inherent flaws in OTD, but rather the fact that OTD was implemented in a world where the loan underwriting systems were flawed. This follows directly from what Comptroller Dugan is asserting.
What I am sure bankers picked up on was the phrase "establish minimum underwriting standards by regulation." In other words, not leaving it up to the lender to decide, but instead to require that certain guiding principles are followed, no matter what the models say! Even if the lender finds "substitutes for common sense" as I called out in my Congressional Testimony, they must still exercise sound judgment in making the loan. I concluded my oral and written testimony with the following statement: "A comprehensive framework that combines the best that judgment and science have to offer, can greatly enhance existing credit scoring models and underwriting processes, and ensure fair access to credit by promoting transparency and common-sense." So, what sort of minimum requirements is Comptroller Dugan thinking about? Well, he goes on to suggest what is needed and that he considers to be "basic, core standards on which there is clearest concensus." On Comptroller Dugan's short list are the following primary lending factors, with the corresponding "C" from the CCAF 5 C's of Credit indicated next to each: • Effective verification of income and financial information; (Capacity and Capital)
• Meaningful down payments; (Collateral)
• Reasonable debt-to-income ratios; (Capacity) and
• For monthly payments that increase over time, qualifying borrowers based on the higher, later rate, rather than the lower, initial rate. (Conditions) The first thing I noticed was the absence of the FICO score as a primary indicator of credit worthiness. Next, I noticed that Comptroller Dugan's short list includes four of the 5 C's of Credit. Relative to the 5th C, Conditions, there is a similar focus to what I have explained in the CCAF white paper, my Congressional Testimony, and the book on the new lending system, namely that borrower vulnerability must be considered by substituting the future values for such quantities as the interest rate on a variable rate loan (which affects the payment amount and hence the debt-to-income ratio); the value of the collateral minus two standard deviations based on five years of history, and also the remaining loan balance at the loan reset date (which exert a combined effect on the loan to value ratio) and so on. Yes, with his short list, Comptroller Dugan has come very close to the baseline for our newly proposed lending framework!
I was particulary gratified to hear his conclusion, which echoed my sentiments about the avoidable nature of the economic crisis, had CCAF been in place. Specifically, Comptroller Dugan observed:
Other standards could be included as well, but the ones I’ve listed strike me as crucial, at least in this sense: had they been in place for the last ten years, I am confident that we would have had nowhere near the problems we have had in the residential mortgage market." Had the new lending system I am proposing been in place in the beginning of this decade, the housing bubble would not have materialized and there would have been no crisis. This is because CCAF would not have approved loans for people who had little means, wanted more home than they could afford, and who were depending on home appreciation to subsidize a higher lifestyle than was sustainable, and who were counting on market liquidity in the event they needed to exit from their mortgage obligation.
 Let me make my professional opinion on the root cause of the financial crisis crystal clear. The crisis was not the fault of the home buyers. The crisis was not the fault of exotic loan products, villianized by many, who referred to them as financial time bombs by some. The key is not putting risky borrowers with risky products, which is what occurred on a mass scale. We drive this point home in Chapter 4 of our book on Borrowers with Exhibit 4.15 from our most recent book, Credit Risk Assessment: The New Lending System for Borrowers, Lenders & Investors, Clark Abrahams & Mingyuan Zhang, copyright 2009, SAS Institute, Inc. (Exhibit 4.15 is reprinted in this blog post with the permission of John Wiley & Sons, Inc.)
As My co-author, Dr. Mingyuan Zhang, and I explain in the first chapter of our second book, a primary cause of the financial crisis was flawed loan underwriting systems, including a huge over-reliance on the FICO score, which was the primary mechanism for pricing consumer credit risk and creation of loan pools to sell to investors. I know this to be a fact, first-hand, because, prior to my move to the technology sector in 2003, I worked in the business for 29 years, built, reviewed, or oversaw the development of hundreds of scorecards for the full range of consumer and small business loan products, and created the FICO-based selection logic for creating loan securitization pools for one of the largest credit grantors in the US.
We identify two additional root causes of the crisis in Chapter 1 of our book on the new lending system, namely lack of transparency and greed. They relate directly to the flawed underwriting systems. Lack of transparency hindered the detection of the underwriting flaws, the loan securitization flaws, and the assumptions behind cash flow models supporting the CDO and other mortgage-backed securities. Greed played in as there was no financial incentive to fix loan underwriting systems that appeared to be so lucritive. In June of 2006, at the ABA national Compliance Convention in Orlando, I warned bankers about the foreclosure trends and asked that they regulate themselves and consider the suitability and affordability of the loan products for different segments of borrowers. I told an overflow crowd of attendees in two sessions that unless they took the appropriate steps to change their loan underwriting practices they would reap the consequences of higher loan losses and greater regulation. I delivered that message three and a half years ago at a time when there was no mitivation to spend money on changing loan underwriting systems that were extremely profitable in the short run, and, as it turns out, terribly flawed at their core. (My presentation was taped and for $20 you can order a copy from the American Banking Association and I also have a white paper that was available to conference attendees, published by SAS if you are interested.) I welcome your comments!
Wednesday, February 3. 2010
The new lending system (CCAF) delivers the power to know. Who would argue against this capability? Well, I had an interesting conversation that I want to share about a pushback on the new lending framework. A privacy advocate was horrified that the CCAF handle (credit qualification classification) would tell someone everything about a consumer's loan qualifications with a single number. My response is simple and direct, "Yes, that is the whole idea!" "Very bad, from a privacy standpoint" is the reply. However CCAF is more accurate, simple, safe and transparent from a credit risk management view than what is in use today. It occurs to me that we seem to have a different mindset about privacy depending upon the circumstances. Let me illustrate with an analogy and then tie back to the current situation, where we have huge credit demand, and far less lending than is needed.
Folks are generally private about their medical condition (e.g. presence of genetic traits, or diseases they suffer from, and so on). However, when placed in a life and death situation, all of a sudden, the patient is willing to let first responders on the scene (even non-medical folks) know everything about them in the hope it will increase their chance of survival. Folks wear medic alert bracelets, so that, even if they are unconscious, the care provider or first responder will know factors that are of primary concern regarding the patient's health. Survival, not privacy, is of paramount concern.
Let’s revisit lending. Default on a mortgage results in foreclosure, physical displacement of an entire family, loss of equity, and destruction of good credit standing for years to come. Those consequences seem to me to be pretty drastic. If having a single number that would preclude the possibility that consumers are put into unaffordable loans, then wouldn't that be of paramount importance? We wouldn't be in the crisis we find ourselves in today if we had a system like what I am advocating in the first place. Furthermore, knowing someone's CCAF handle would not equate to knowing their social security number or driver's license number - in other words it is not unique to a particular individual. Identical protections that are currently in place for the credit bureau score could be deployed for the CCAF handle and score. The primary difference between the CCAF handle and the credit bureau score is that the CCAF is more comprehensive and also the fact that, although it is a single 3 digit number, it can actually be decomposed into credit ratings for credit history, capacity, capital, and collateral strength and/or weakness and also vulnerability of the borrower to future conditions based on the type of loan. The bottom line is that the CCAF handle is packed with far more information than the FICO credit bureau score.
Society can't have it both ways. If gathering data on consumers, and developing powerful ways to summarize that data is viewed as a threat, then we can either opt to not leverage data and operate more in the dark, or we can prohibit the sharing of information and operate with less transparency. In my opinion, both of these paths would be a mistake.
SAS CEO Jim Goodnight put it best, and succinctly, in a piece considering the use of alternative data in lending some time ago when he advised "Let the data speak." Operating in the dark is not a helpful strategy! On the transparency front, we need more, not less of it. That is a lesson coming out of the financial crisis. Further, we need the lender, borrower, loan servicer, packager of loans, and finally the investor in asset-backed securities (ABS) to all have the same transparent and consistent view of loans.
That is precisely what CCAF delivers - a consumer credit qualification based on more data, a more integrated view through the data, and a single transparent number that provides all interested parties with the relevant information they need to make the best informed decisions. I sincerely hope that reason prevails, and that the privacy advocates will embrace, rather than oppose, the new lending system. To do otherwise would hurt the very people whose interests they aspire to represent -- i.e. privacy intact, financial affairs in ruins!
Sunday, January 24. 2010
I read David Lightman's piece yesterday in the Raleigh News & Observer about the long shadow on the Fed Chairman's confirmation that is cast by The Wall . That's right, I said "The Wall," and by that I am talking about the unnessary obstruction that is effectively separating borrowers and lenders; The Wall that is preventing the flow of money to consumers and small businesses that is so vital to our nation's economic recovery. Lightman reported that, in reference to a meeting Senate Majority Leader Harry Reid had last Thursday with the Fed Chairman, Reid said: "I made it clear that to merit confirmation, Chairman Bernanke must redouble his efforts to ensure families can access credit they need to buy or keep their home, send their children to college or start a small business." The problem is that we are stuck in the business as usual old "FICO-based" way of thinking about credit granting and too reliant on the credit bureaus to "score" all of what is important in loan decisioning. What are lenders to do if the gauges on their credit risk meters are fogging up with unreliable scores and conflicting information that is hindering their ability to determine how risky loan applications really are? Answer: a score refresh, a model re-callibration, and pulling levers based on individual dial readings simply will not suffice. They need only one integrated dial that sums up all of the relevent information and represents a single version of the truth. Not the way lending has been done that led us to where we are today, but rather the reality of what it needs to bercome. What's needed here is: METHODOLOGY REGIME CHANGE! The Fed Chairman needs to stop letting other folks who have a vested interest in preserving the Lending Status Quo dictate what our options are and effectively stone-wall calls for change. In short, Ben Bernanke needs to say "Enough!"
In his January 20 speech, entitled " Lessons of the Crisis: The Implications for Regulatory Reform," William C. Dudley, President and Chief Executive Officer of the Federal Reserve Bank of New York, said: "We need a new regulatory structure that provides for comprehensive and consistent oversight of all elements of the financial system." FRBNY CEO Dudley goes on to say: "No economy can prosper without a well-functioning financial system—one that efficiently channels savings to the businesses that can make the most productive use of those savings, and to consumers that need credit to buy a home and support a family." I would argue that strenghtening the banking system is not so much a matter of requiring more capital to support lending operations (as many have suggested) -- rather reform should be more concerned with ensuring that lending practices are safe and sound in the first place, in which case perhaps less capital is needed! But, you may wonder, "How is this to be acomplished?"
The good news is that the Fed Chairman does not need to invent the solution to providing greater and saver access to loans for consumers and small business owners - SAS already has -- it's called the Comprehensive Credit Assessment Framework and it is The New Lending System for Borrowers, Lenders, and Investors! How does this new system improve upon the current "state of the art?" Answer:1) For starters the holistic CCAF score for a loan applicant relates to all of the borrower and transaction information, not just what is in their credit file! That's right, the CCAF is the only score that considers everything relevant to the lending decision.
2) Unlike credit scoring, the CCAF score does not impose greater weight on one primary factor (i.e. credit payment history) versus any others (i.e. character, capacity, capital, collateral, and loan terms and current and future economic conditions). Credit scoring inherently suffers from this weakness, due to its core, marginal-contribution-based, development methodology, which it persists through the entire life of the non-adaptive, one-size-fits-all scorecard.
3) The CCAF system associates a classification (called a handle) with every loan application that provides in-depth information on how the borrower rates relative to the primary loan qualification factors that make up the holistic CCAF score. In that way, borrowers and lenders are "re-connected" by sharing the identical transparent view.
4) No more secret sauce FICO score. Essentially, lenders who use the FICO score are saying to the borrower "We don't want you to have the power to know:
1) how you are evaluated and what weighs in importance in the decision to grant the loan, and/or
2) how much to charge you because of how much risk our models tell us you represent!" That doesn't exactly foster trust in the financial system -- it only serves to reinforce The Wall between borrowers, lenders, and investors. Yes, The Wall that stands between consumers and their unmet needs; make no mistake, I'm talking about The Wall that small business owners must scale in order to survive. Tearing down The Wall should be Fed Chairman Bernanke's number one goal if he truly hopes to achieve a sustainable economic recovery. Yes, indeed, the chorous from consumers, small business owners, and lawmakers is clear, and growing louder every day,and I invite you to join me in urging the Fed to push for the type of change that's needed, to the rallying cry: Chairman Bernanke, Tear Down That Wall!
Friday, January 22. 2010
 Yesterday, I learned that my employer, SAS Institute, was ranked #1 on the FORTUNE '100 Best Companies to Work For' 2010 list. This is big news to outsiders, but no surprise to me. SAS Knows That It's All About People.
You see, I was a customer of SAS for two decades prior to joining the company late in the Spring of 2003. At that time I had already come to know SAS as a company that would stop at nothing to help me achieve success in meeting my business goals. Indeed, my career in banking really accelerated when I joined what is now Bank of America, and began to apply the SAS technology to solving very complex business problems. Those business applications ranged from sourcing and validating data across the enterprise, to pricing derivative securities, to balance sheet strategy development using large-scale financial optimization, to credit risk management, and finally finding lending opportunities in underserved communities and ensuring that lending practices were fair and responsible. Armed with "The Power to Know" I rose through the ranks form AVP to VP to SVP and finally capped my career as a CRO reporting to the BOD of a community bank based in Atlanta, Georgia. The insight gained through the use of SAS solutions and tools provided the competitive edge that made all of the difference in my banking career.
 As a member of the SAS Team for nearly seven years, I have cultivated an even greater appreciation of what SAS is all about, namely, PEOPLE! This is a company that goes to great lengths to create a work environment that focuses on meeting the needs of individuals so that they can, in turn, maximize their contribution to helping achieve the corporate goals. The company recognizes the importance of balancing work and life outside of work. Jim Goodnight has molded a culture that encourages new thinking, respects and values the individual, and promotes collaboration in the development of intellectual capital (i.e. folks share ideas and co-invent new and better ways to do things). I know this first-hand, having co-authored two books in the SAS/Wiley Business Series on lending, being an inventor and co-inventor on a half dozen patents, and by collaborating on a solution that will help address the root causes of the biggest financial problem our nation has faced since the Great Depression! Jim's leadership inspires innovation and promotes inclusion. We at SAS are a world-wide community of professionals, who are dedicated to delivering value to our customers -- value that is rooted in the best that leading-edge science, coupled with expert business judgment, has to offer.
It is a privilege to work in such an incredible environment, and I am proud of what SAS stands for: The Power to Know
Power to Its People
Yes, indeed, SAS Knows That It's All About People.
Tuesday, January 19. 2010
I want to do some myth-busting around the notions that: 1. Poor people are risky
2. Cash payers are risky
To make it real, I want to share a lending story about a woman who is small in stature and had very meager beginnings, but who is a giant in terms of faith and determination, Her name is Estella, and I actually dedicated both of my books to her and she has been a lifeling mentor to me. Estella was raised in poverty in rural Tennessee, and her parents were share croppers. She had to overcome many hurdles involving economic hardship and access to education.
Estella's life story is an inspiration, but the focus here is the challenge she faced to qualify for a loan. Those were the days when credit scoring was dawning at Fair Isaac and Company, but scoring's use did not become more widespread until a decade later, and almost exclusively for credit cards, and some installment lending, but not mortgage lending. Make no mistake, when she applied for a loan, Estella was more than likely evaluated using the long standing and widely accepted principles of the 5 C's of Credit, namely Character, Capacity, Capital, Collateral, and Conditions. Furthermore, Estella was not reduced to a number, rather she recounted how she sat in the branch, face-to-face with a loan officer, with whom she shared her story, her faith and determination, her qualifications, and her dream of owning her own home.
I first met Estella in 1959 when my parents employed her to help out in the home. In 1961, at age 47, Estella decided to take a big risk and buy a home. Everyone advised her against it, saying it was not a good idea, and that in those days an unmarried woman, especially being black and older, had too many forces working against her. But they did not know Estella and how much she wanted that home, which she saw as her security in old age and a chance to own a piece of the community in which she had decided to live. Estella possessed deep faith, and when she looked for a place to live, she called on The Lord for help. How she knew she had the right home is another story by itself, but suffice it to say she found the right one. It had a nice yard in the back and was close to a bus stop.
Estella always paid cash for everything. She had no record in the credit bureau, no formal education, no permanent job, but steady domestic work for several families and no debt. Her wage was a dollar an hour and she had to pay her own way to get to and from work. Her life savings of $4,700 made up the 20% down payment on a $21,000 duplex home in San Francisco, leaving no financial cushion. She applied for a 15 year fixed rate loan. Here’s how she rated at loan closing: her persistence, strength and faith were all very strong! Her other ratings were: credit history – none, capacity – poor, capital–poor, collateral – good, conditions – good. Based on judgement, the lender decided to take a chance on Estella and she got the loan!
Back then, I recall Estella telling me how she stood in line to be the first customer when Wells Fargo Bank, the same bank that granted her her home loan, decided to open a new branch in her neighborhood. Of course, Estella was first customer in line on that day. Her reward was a checking account with free checks! She also opened a savings account that day, and every week, without fail, she would go to the bank and deposit her savings for the week. Sometimes she would have only loose change amounting to less than a dollar, but the teller always thanked her for her deposit and updated her passbook. Estella grew many of her own vegetables in her backyard and she did not waste anything. She even made her own soap and stain remover with leftover grease she would save in jars, some lye, and other ingredients. In fact, I still have two white bricks from a batch she made 10 years ago that can take the spots out of my white shirts better than any commercial cleaner! Estella knew how to do everything, and she could "make do" with almost nothing.
Fast forward:
In 1974, Estella paid her mortgage loan off in full (two years early at age 60).
In 2005, Estella sold her home for $1.4 million, with zero debt.
Today Estella is a millionaire at age 95.
I say that we need to give the Estellas of the world a chance. With CCAF, this is possible. With today's "business as usual," FICO-based lending systems, it's simply too steep a hill for borrowers to climb. I challenge you to ask yourself these four questions: 1. Do you think a FICO score would be an accurate measure of Estella's creditworthiness? (Assuming they could even calculate one for her)
2. Do you think the commonplace FICO-rooted lending systems in use today are best serving the needs of creditworthy consumers?
3. Do you think it is appropriate for lenders to use secret sauce measures of creditworthiness like the FICO score insead of a more transparent system like CCAF that provides the same view to borrowers and lenders alike?
4. Do you think the commonplace FICO-rooted lending systems in use today foster fair and responsible lending?
Thursday, January 14. 2010
Karen Blumenthal, Wall Street Journal writer, wrote a great piece yesterday entitled " Look Who's Peeking at Your Paycheck ." Ms. Blumenthal reported that lenders will be gathering more data on consumers, such as income, assets and current debt level. In fact, the Fed is requiring that lenders obtain and use this information in their lending operation.
This is precisely what we have been advocating, both here and abroad, for the past two years. We published a white paper detailing the need for this in early 2008, and in the Summer of 2008 I made the same case during my Congressional Testimony “What Borrowers Need to Know About Credit Scoring Models and Credit Scores”(carried live on CSPAN-3).
It is the subject of our latest book, published in the Spring of 2009. In that book, we illustrate how borrowers can benefit from a new lending system that is a comprehensive credit assessment framework (CCAF, pronounced See-Caf).
I encourage every consumer to read chapter four of that book to learn the truth of the matter -- not ads they see on TV, or the Internet! The book has been added to the collections at several public and university libraries and you should check with your local library for availability.
Wednesday, January 13. 2010
In the first chapter of my latest book, Credit Risk Assessment: The New Lending System for Borrowers, Lenders, and Investors, I quote John Dewey, educational pioneer, who said "A problem well-defined is half solved." The key to solving the loan underwriting problem that exists today is framing it -- that is, identifying the primary factors that should be taken into account, jointly, when qualifying a prospective borrower for a loan. The good news is that there is no need to "reinvent the wheel." We can tap the 5 C's of Credit, which far preceeded the development of credit scoring methods. Simultaneously considering all of the primary factors enables us to: PUT THE PROBLEM INTO THE PROPER CONTEXT! That is what CCAF accomplishes. So, you may wonder, what is CCAF and how does it compare with the FICO credit bureau score?
To answer the first question, there are several sources that explain CCAF. I have provided the basics in prior posts to this blog over the past year. There is also a white paper, and two books. The first book has a more technical treatment in the last three chapters, and the second book provides a very detailed business-oriented description in Chapter 2, and then examines it from the lender, borrower, investor, and regulatory perspectives separately in Chapters 3-6, respectively.
To answer the second question, I constructed a table that lists some key features and compares the two approaches side-by-side.
Relative to the financial crisis, let me make a few observations concerning consumer lending:
Observation #1: loan underwriting models have evolved to be piecemeal; the value chain is “riddled with disconnects”
Observation #2: FICO scores are no longer as predictive as in the past
Observation #3: We need better early warning capabilities that can effectively spot trends that foreshadow loan performance declines
Fact: Behavioral patterns may be precursors of loan defalult for some, and "pure noise" for others. The key is to put those patterns into the proper context!
Fact: The FICO score is an incomplete measure of borrower credit risk and is not reliable!
Fact: FICO credit scoring systems are “Black Boxes” and they are hard to interpret: Can’t look at a scorecard and tell if it is correct!
Fact: FICO Credit scoring is a “one size fits all” approach – everyone is asked the same questions and constant points are assigned to responses, regardless of answers to prior questions: There is no “average” borrower, hence every scorecard factor is taken out of context!
Fact: Scorecards are not adaptive and their predictive power diminishes steadily over time: Scorecards developed today may only have 6 months of useful life by the time they are developed and installed!
Fact: Scorecard validation from holdout samples of loans made during identical timeframes and conditions assumes that historical correlations will persist under future undetermined scenarios without any structural or common sense support: The validation process for scoring models is too narrow in scope and lacks common sense tests!
Fact: Credit scoring seeks to risk rate first, based on a subset of factors, and then classify borrowers by score ranges: Borrowers should first be classified based on all relevant factors and then risk rated!
Fact: It can be difficult to balance compliance requirements with credit scoring, which is subject to overrides: Borrowers may have “identical FICO scores,” but vastly different qualifications!
Today’s underwriting, loan pricing, and loan securitization systems rely too heavily on the FICO score and credit bureau data. Greater emphasis should be placed on capacity, capital, collateral, and conditions of the loan, the borrower and the economy. Are we to believe that all of the consumers who are finding themselves in financial trouble, possibly foreclosure, are in thier position to lack of character? Or could it be due to the conditions, combined with thier capacity, capital, and the volatility of the value of their collateral that put them in a position where their loan(s) became unaffordable?
In past blogs I have used plenty of medical analogies, so, for a change, consider the following driving analogy. Suppose one out of every five cars of a given make and model coming off the assembly line ended up triggering a fatal accident. Would you say those accidents were a result of driver error?
If we are going to hold up a score that is supposed to indicate someone's creditworthiness, shouldn't that score be based on all of the relevant factors? And to think that the FICO credit score is also used for employment purposes, insurance qualification, and, who knows what else?
I conceed that the FICO score many be a statistical summarization of a lot of what is in your credit bureau file. However, as I have maintained throughout, it is all a matter of context -- precisely what the FICO score lacks!
Monday, January 11. 2010
I want to share with you the latest US-based push-back I’ve heard about CCAF. It has to do with Regulation B, The Equal Credit Opportunity Act. A banker has said that CCAF will never be embraced by the lending community until the regulators can approve its methodology for specifying the reasons for decline when a credit application is rejected. Furthermore, since CCAF is not “vanilla credit scoring,” there is some question whether or not it would be considered by law to be a judgmental system, as opposed to a scoring system.
In this blogpost, I will probe the subject of ECOA decline reasons a bit and illustrate why actually it is the lenders using the FICO score that should be concerned, and not those who contemplate switching over to a more Comprehensive Credit Assessment Framework ( CCAF for short, pronounced See-Caf). It should be readily apparent to all by now why the FICO-based loan underwriting status quo served neither the lenders nor the borrowers well as we gradually slid over the years into what has emerged as a huge financial crisis.
The time has come for greater transparency and disclosure. That is one of the principal reasons why CCAF was developed, i.e. to address exactly those weaknesses in the credit system. Instead of just telling borrowers, plain and simple, exactly what factors are used to evaluate, and score, them, they are instead given "clues" via Regulation B notifications. I have heard it said by the proponents of the FICO system, even in the Congressional Hearings in which I testified in 2008, that the "secret sauce" is necessary, and that we cannot trust the consumer not to try to "game the system" if FICO and the Credit Bureaus told "them" the rules.
My response is "Who's gaming who?!" Trust goes both ways. It is high time that we afford the American Public with the information and respect and trust that they deserve. That is what CCAF will deliver - everything out on the table, the way it should be. And as for fraud, the technology exists to authenticate individuals, and, if necessary, verify everything that they report is accurate. (In a later blog, I will return to the advantages of co-deployment of fraud and credit frameworks.)
A Bit of Background
I was working in consumer credit at AMOCO in 1974 when the Equal Credit Opportunity Act was enacted. Essentially, it specifically prohibited lenders from discriminating against a loan applicant on the basis of gender or marital status. Later, in 1976, Congress passed amendments to ECOA, and effective March 23, 1977, race, color, religion, national origin, age, dependence on public assistance or exercise of rights under the Consumer Protection Act were added as prohibited bases. Three weeks later, on April 14, 1977, I addressed lawyers and lending executives from 35 states and Canada on issues surrounding ECOA (and its most recent amendments) at a conference held at the Fairmont Hotel atop Nob Hill in my home town of San Francisco. I later served as a panel member with William Fair, founder of Fair, Isaac, & Company (recently re-branded as FICO). Yes, my formalized study and comments on Regulation B date back 33 years, and it has remained an area of great personal interest to me. I performed analysis on extensive data relating to lawsuits against lenders alleged violations of ECOA both at FICO and later as an expert witness as an independent credit management consultant.
Back to Reasons for Decline
In ECOA, Congress required that the reasons given must be specific, but it left the door open as to what constitutes specific reasons for credit refusal by a scoring system. A scorecard does not pave the way for a lender to make an accurate and useful identification of reasons for an application's refusal. Each applicant's score is derived from the totality of factors included in the scorecard. In point scoring models, points are awarded for each scored factored and then totaled. Applicants are rejected when their total score falls below the prevailing scoring system cut-off. It can be argued, that any factor where the applicant does not receive the maximum number of points is a factor that contributed to the decision to decline credit. Complicating matters is the fact that lenders adjust their cut-off score over time, which means that some applicants that are rejected, would in fact have qualified if only they had applied at a different time. That said, it was the intent of Congress that lenders provide rejected applicants sufficiently specific reasons for adverse action so that they could use the information to take actions that would result in an improvement in their score. A distinguishing feature of CCAF over credit scoring is common sense and the use of proven credit principles. In my mind, that is an advantage, not a disadvantage!
Let’s consider some cases.
Tuesday, December 22. 2009
Lending is a noble profession. It is integral to the operation of a free society and free markets. If we are to achieve an inclusive society, we must foster economic inclusion. That is the spirit behind the CRA legislation, passed in 1977. Despite allegations from misinformed individuals, CRA had no more to blame for the financial crisis than its predecessor, the Fair Housing Act of 1968 which prohibited discrimination in housing.
Consider that the basic principles of lending have not changed in 1,000 years. What did change in the 1960's was the scope and scale of consumer lending, fueled by the introduction of the credit card and its adoption by financial institutions, oil companies, and department stores. Lenders sought a way to decision credit card applications more rapidly, and credit scoring models enabled them to keep pace, and its usage spread to other loan products.
Fast forward to today, and the financial crisis. In case you missed past blogs and a couple of books noted in the righthand margin of my blogspace, I have been making the case that we could have avoided the mess if we had put in place a comprehensive credit assessment framework ( CCAF, pronounced See-Caf).
Bold claim? Maybe, but I back it up with an in-depth description of what a comprehensive credit assesswmnt framework looks like, how it operates, and I document its clear superiority to today's fragmented and FICO-reliant underwriting systems. It fixes the problems associated with the root causes of the crisis, which we identify in our latest book: Credit Risk Assessment: The New Lending System for Borrowers, Lenders, and Investors, namely: 1) greed
2) lack of transparency
3) flawed risk management. How does CCAF address thes root causes? Answer - systematically and comprehensively! Let me elaborate on how point #3 relates to the first two.
Monday, December 14. 2009
Today, President Obama is meeting with executives from the largest banks in the country in a effort to increase lending to consumers and small businesses. This is going to be an uphill conversation for our President, because lenders are continuing to tighten their requirements for these two sets of borrowers (see The October 2009 Senior Loan Officer Opinion Survey on Bank Lending Practices). In last week's blog, I noted a recent Washington Post article that gave some specifics on tighter standards that Fannie Mae just put in place on December 12 in the mortgage lending area! I want to circle back to that point in a minute.
In yesterday's New York Times article by David Streitfield, entitled Mortgage Rates Lowest Since 1940's, But Fewer Qualify, it was reported that mortgage rates in the US have dropped to their lowest levels in over 60 years thanks to the trillion-dollar capital infusion by the Fed. As David Strietfield points out, the current scarcity of credit not only hurts homeowners, 60 percent of whom pay higher than the current 4.8% 30-year fixed mortgage rate on their existing home loan, but also has broad economic repercussions. Consumers and small businesses find themselves in a tough position today. What good is it if rates are at a historical low, but borrowers are unable to take advantage of that fact?
What about the bankers? Well, they also find themselves in a tough position. On the one hand, they are trying to maintain greater capital reserves (at the urging of their primary regulator) and improve asset quality (which breeds extra caution in lending). On the other hand, they are being urged by consumers, and now the President, to losen thier credit policies, which takes us back to last week's blog post. Try to imagine how banks can effectively loosen their mortgage lending requirements, while Fannie Mae (a quasi-governmental agency that controls loan eligibility for pools of mortgages that back securities sold to investors in the capital markets) is now tightening theirs? The reality is that banks can only hold so many mortgages on thier balance sheet, and Fannie provides lenders with the ability to originate a far greater number of loans that can be pooled, pakaged, and sold in the capital markets. Does this sound like mixed messages for the bankers to attempt to sort out?! Let me know what you think!
With a comprehensive credit assessment framework, it would be possible to satisfy all of the aforementioned objectives. Indeed, this new lending system for borrowers, lenders and investors is exactly what President Obama is looking for to safely, and significantly, get the money flowing again to qualified consumers and small businesses. If adopted universally, it could help us to get this country's economy back on its feet again.
What we need now, more than ever, is bold and decisive action. What we don't need is more of the status quo, with its continued emphasis on pulling individual credit levers that unnecessarily disadvantage, at once, entire segments of creditworthy borrowers, and the over emphasis of the FICO score, which has experienced huge downward migration over the past year for millions of consumers, even in the super-prime segment. Yet we continue to hear proponents say that the FICO credit bureau score is predictive, that it validates and that it rank-orders risk! That is simply garbage!
First, the magnitude of FICO score migrations show it to be a volitile measure of risk, and, as such, not terribly predictive or useful. Second, scores that validate with historical samples do not inspire much confidence when past assumptions no longer hold and historical data do not reflect what is actually happening now. Third, one has to wonder how can a FICO credit bureau score rank order risk when it does not consider borrower capacity, capital, or the collateral and loan terms? That is like saying that a patient's family medical history rank orders the risk of disease (pick one) in the absence of physical examination, blood tests, x-rays, a biopsy, an ultrasound, and so on. When all relevant information is considered, a rank ordering makes sense. When only a norrow set of information is considered, it amounts to a false and misleading claim. The bottom line (as Dr. Zhang and I detailed in our latest book) is: lenders continue to rely on the same kind of incomplete and flawed information that got us into this mess in the first place.
So far, the focus of this blog post has been on loan underwriting, but I simply can't resist the urge to go a step or two beyond to point out the operational risk exposure that is ever-present throughout the lending value chain when one uses the FICO credit bureau score as a measure of loan default risk. Talk about "out of context," are we really supposed to believe that a score that does not consider how much a borrower makes or how much a borrower has managed to save can actually predict the likelihood that they will repay the loan? My advice to those of you who use FICO score buckets to allocate capital, price loans for sale in the capital markets, or to segment portfolios for loss forecasting purposes, you would be far bettter served to adopt the CCAF handle in its place for greater: 1) accuracy,
2) consistency,
3) transparency,
4) interpretability,
5) completeness,
6) adaptability, and
7) flexibility. Continued reliance on FICO score buckets for the aforementioned purposes is to embrace a methodology that: 1) loses accuracy the minute a key linkage changes, which you most likely will not detect until it is too late to do anything about it,
2) inherently fosters inconsistency due to the inevitable and significant score overrides,
3) is a black box, affectionately termed secret sauce,
4) utilizes substitutes that, while statistically correlated, fail the test of reason,
5) fails to include entire categories of relevant information, in fact 4 out of the 5 C's of Credit are absent,
6) is static and whose accuracy diminishes steadily over time, and
7) is inflexible with a fixed number of factors and point assignments that, like the boy with a hammer, treats every consumer like a nail.
Do you think we need change now? Please post a comment. I am very interested to know what you have to say about these immensely important issues.
Friday, December 4. 2009
Did you happen to see the Washington Post article a week ago saying that Fannie Mae is tightening its lending standards? That article was of significant interest to me, because I have been describing to you the challenges for today's lenders in tightening lending policies in the absence of having a comprehensive credit assessment framework (CCAF) . Please check out my blogpost on Sunday, October 4 for more details contrasting CCAF's approach with the more blunt and opaque solutions offered by underwriting systems that utilize the FICO score.
According to staff writer Dina ElBoghdady, "Starting Dec. 12, the automated system that Fannie Mae uses to approve loans will reject borrowers who have at least a 20 percent down payment but whose credit scores fall below 620 out of 850. Previously, the cut-off was 580." One can only wonder whether or not that makes sense for borrowers who put 40% down. Or, how about borrowers with 24 months of liquid reserves? The FICO score is really focused only on tradeline payment performance, and hence is an incomplete measure of risk. Thus, The FICO score can be very mis-leading when used in isolation. Changing the rules by raising the floor on the FICO credit bureau score is ignoring such things as borrower income, capital, equity position and terms and conditions of the loan. Reading further, Dina goes on to say that: "Also, for borrowers with a 20 percent down payment, no more than 45 percent of their gross monthly income can go toward paying debts. Fannie declined to disclose the previous threshold, except to say that it was higher. The company will raise the level to 50 percent in cases with "strong compensating factors." Based on my reading of the article, it appears that Fannie Mae has developed a collection of individual rules that are applied, with the end result being that certain segments of borrowers, who may indeed be well qualified by virtue of the other C's of credit , will nonetheless be over penalized. Clearly, what they could really benefit from is a holistic framework that considers all relevant factors simultaneously.
Perhaps a simple example, or two, will drive home the point. My son just graduated from medical school and is interning prior to starting his 4 year residency in radiology. He and his wife, a 4th year medical student doing rotations in several different cities and states, both need cars. The housing market will likely be at bottom next summer when he moves to do his 4 year residency and buying a home would make good financial sense. However, based on these rules, and his student loans, and their car loans, he won't qualify by Fannie's standards, and will probably live in an apartment with wife, dog, and possibly children, for the next 5 years. Does that sound right?
Or, suppose someone who has saved up a 20% down payment on a modest home, has an additional 9 months worth of liquid reserves, and makes a modest income driving a MUNI bus for the city of San Francisco. Suppose he/she is willing to sacrifice lifestyle to have a home near where he/she was raised, but where housing is very expensive. Right now may be the perfect opportunity to get in at the bottom - but because of the absolute 45% - 50% cap on debt ratio the bus driver does not qualify. The home could be his/her security in old age, and a path to greater wealth along the way. Does that make any sense?
The net result of this type of approach to credit tightening, or rationing, is that it serves to widen the gap between risk management and credit access, which was the topic of my keynote speech in Cartegena, Colombia the end of November, that the previous blog explained was so well-received. How is it that they "get the CCAF message" in other countries, but it doesn't seem to be taking hold where we need it to -- right here in the US where it was developed?!
I don't mean to pick on Fannie Mae. In fact, I have great respect for the people there and their best intentions to do what is right. In fact, this is a systemic problem that I am addressing. What we all need to do is to break away from the old ways of thinking, and out-of-date constructs like the FICO bureau score, which is not reflective of anyone's true credit worthiness. And, oh by the way, investors in mortgage-backed securities need transparency down to the loan level in the investor pools that can be best accomplished through the handle construct of CCAF, and not simply the ratings offerred by the major rating agencies that have, like the FICO bureau score, issues about thier reliability based on actual performance.
As long as lenders continue to make policy decisions in the typical fashion that prevails in the market today, they will continue to fail to achieve thier goals and the borrowing public will continue to be punished unnecessarily. As a result, the markets will fail to achieve the volume and liquidity that it so desperately needs to recapture, and a quicker economic recovery will reamin elusive.
Business as usual is not an option. Rather, as I noted in My September 14th blog post, it is a prescription for extinction.
Wednesday, December 2. 2009
 Wow, we had a wonderful response on November 20 in Cartegena, Colombia and many follow-up conversations with attendees about the virtues of CCAF and how it can help close the gap between credit access and risk management. Unfortunately, I came down with the flu several days prior to my scheduled trip. Fortunately for everyone, my colleague Jose Etchegoyen, in the SAS Risk Practice, delivered the speech in my place, and he not only saved the day, but did an outstanding job with only last minute notice.
A giant THANK YOU to Jose !!
Keynote topical outline: 1. Introduction
2. Credit Access and Technology’s Role
3. The New Lending System
4. Transaction Risk Exposures/Missed Opportunities
5. Big Picture View: Product Lines/Institutions/Countries
6. Summary 
In a future blog I will provide information on where you can get a copy of the speech. In the meantime, I need to coordiante with the conference sponsor on their plans to publish the proceedings. The message is very timely and applies to lenders and borrowers residing in any country in the world!
Saturday, November 14. 2009
I have literally been on the go for the past couple of weeks getting the word out on CCAF in five cities, the last of which was Orlando, where I spent several days attending RMA’s Annual Conference on Risk Management. Loan underwriting was the focus of much of the conversation, together with upcoming regulatory reforms and the economic outlook, at both the national and state levels.
Next week finds me headed to South America. So, I wanted to catch-up before too much time passed by to let you know what I am hearing and thinking about!
In the community development space, I heard some exciting news while attending a Social Compact Board Meeting. CEO John Talmage shared that Manny Diaz, the outgoing mayor of Miami, has credited their DrillDown as possibly the most significant project of his administration (Social Compact’s DrillDown is powered by SAS technology). The DrillDown of 14 Miami neighborhoods showed that the city has 137,000 more people than the Census projected, as well as a 29% higher income level. At present, Social Compact is also in partnership with the World Bank and local officials to test the potential of DrillDown methodologies in emerging markets in Johannesburg, South Africa and Bogota, Columbia.
Speaking of Colombia, and the credit underwriting space, on November 20, I will actually be delivering a keynote address there at a Risk Management Conference sponsored by Asobancaria, the Colombian Banking Association. The title of my presentation is: Credit Access and Risk Management: A Gap To Close. I am looking forward to giving this speech for two main reasons. First is the opportunity to illustrate how CCAF can transform the way loan underwriting is performed today to allow for greater access to credit for consumers falling outside of the financial mainstream. This new lending system is described in mathematical detail in the last half of my first, and technically-oriented, book, published last year. It is the entire subject of my second, business-oriented, book, published this past spring. The second reason for my excitement, and no less than the first, was the opportunity to visit Colombia and learn first-hand about banker’s key concerns and challenges, and to exchange ideas and perspectives.
You may be wondering what information-led community development and loan underwriting have to do with one another. First, they both benefit from having greater information. In the case of community development, more information can be used to develop greater intelligence that can change the perspective of a community from being one of need to being one of opportunity. Loan underwriting can benefit from the sourcing and use of alternative data that can qualify consumers for a loan on their record of making timely cash payments for rent, utilities, phone, and so on. Technology can collect, integrate and organize this information and put it into a form that is amenable for analysis. It can help us to look deeper, below the surface statistics and typical impressions. Of course, in a more direct way, loan underwriting can provide small businesses and consumers with capital that can enable them to fund necessary purchases and, ultimately improve their financial health and overall economic well-being. That, in turn, has a very positive effect on the community in which they are located.
In my talk on closing the gap between credit access and risk management, I explore the connection between community development and loan underwriting in greater depth. Upon my return, I will post what I learn on my trip to Colombia, and I will share reactions to what I have to say about this important and timely topic.
Stay tuned for more on this subject!
Thursday, October 22. 2009
SAS has been conducting research on analytic frameworks that combine expert judgment with best science. We have observed that effective integration of comprehensive views, especially customer-centric ones, with financial performance and the evolving economic and market realities, offers significant competitive advantages.
SAS has actually pioneered a new lending system that, compared with today’s typical loan underwriting systems, is: 1) simpler, yet more accurate,
2) thorough, yet concise,
3) becomes more, not less, predictive over time, and
4) is statistical in nature, yet is transparent and passes various tests of reason by simple inspection. Moreover, these advances in loan underwriting approaches have the implications for the entire lending value chain.
CCAF (pronounced See-Caf) is short for Comprehensive Credit Assessment Framework. It is a new lending system that affords many benefits, in addition to those stated above, when compared with more traditional loan underwriting systems. Additional benefits include: ease of understanding
stresses loan affordability
adaptibility and ease of updating
ability to be effectively monitored
is based on proven lending principles
provides adequate controls to limit risk
affords early warning in advance of performance declines For those who want to learn more, just click on the links in this post to download a copy of the CCAF white paper or to obtain a copy of our book that describes the new lending system. Please share your comments on either, or on this or past blog content. I want to hear your opinions!
Monday, October 19. 2009
I call fact on that! I am in the process of preparing a keynote address entitled Credit Access And Risk Management: A Gap To Close, for an International Banking Conference sponsored by Asobancaria, Colombia's Banking and Financial Entities Association. The conference will take place in late November in Cartagena, Colombia.
During the course of my research on the high cost of being poor in Latin and South America, I learned that it is estimated that perhaps a billion dollars is maintained under mattresses and in cookie jars where it earns no interest and provides no indication of the thriftiness of the lower income tiers of society. Worse, as many as 85 percent of the population is out of the formal financial banking system, and as a result their in-person cash payment and deposit transactions require long trips, risk of robbery, and long waiting lines, in addition to carrying a high degree of cost (as much as five to ten percent of the actual payment amount for check-cashing, and as much as nineteen percent for US dollar international wire transfers to non-bank customers who receive income supplements from relatives and friends). Bank loans can carry as much as a 17 percent annual interest rate (35-40 percent for credit cards) and the common non-bank alternative sources for loans can charge as much as from 150 to 400 percent interest annually.
Continue reading "Widening Credit Access Without Taking On Greater Risk: Fact or Fiction?"
|
Clark Abrahams is Chief Financial Architect at SAS where he leads business and product development. He has over 30 years of experience in the financial services industry. Along with co-author Mingyuan Zhang, Clark has written two books that re-think credit risk management and granting access to credit: 
