In a recent presentation conducted by The Tower Group, “2010 Top 10 Business Drivers, Strategic Responses, and IT Initiatives in Bank Cards,” the conversation covered many of the challenges facing the credit card business in 2010.  When discussing the shift from “what it was," to “what it is now” for many issues in the card industry, one specific point caught my attention – the perception of unused credit lines – and the change in approach from lenders encouraging balance load-up to the perception that unused credit lines now represent unknown vulnerability to lenders. Using market intelligence assets at Experian, I thought I would take a closer look at some of the corresponding data credit score profile trends to see what color I could add to this insight. Here is what I found: • Total unused bankcard limits have decreased by $750 billion from Q3 2008 to Q3 2009 • By risk segment, the largest decline in unused limits has been within the VantageScore® credit score A consumer – the super prime consumer – where unused limits have dropped by $420 billion • More than 82 percent of unused limits reside with VantageScore® credit score A and B consumers – the super-prime and prime consumer segments So what does this mean to risk management today? If you subscribe to the approach that unused limits now represent unknown vulnerability, then this exposure does not reside with traditional “risky” consumers, rather it resides with consumers usually considered to be the least risky. So this is good news, right? Well, maybe not. Vintage analysis of recent credit trends shows that vulnerability within the top score tiers might represent more risk than one would suspect. Delinquency trends for VantageScore® credit score A and B consumers within recent vintages (2006 through 2008) show deteriorating rates of delinquency from each year’s vintage to the next. Despite a shift in loan origination volumes towards this group, the performance of recent prime and super-prime originations shows deterioration and underperformance against historical patterns. If The Tower Group’s read on the market is correct, and unused credit now represents vulnerability and not opportunity, it would be wise for lenders to reconsider where and how yesterday’s opportunity has become today’s risk.  

By: Kari Michel   Lenders are looking for ways to improve their collections strategy as they continue to deal with unprecedented consumer debt, significant increases in delinquency, charge-off rates and unemployment and, declining collectability on accounts. Improve collections To maximize recovered dollars while minimizing collections costs and resources, new collections strategies are a must. The standard assembly line “bucket” approach to collection treatment no longer works because lenders can not afford the inefficiencies and costs of working each account equally without any intelligence around likelihood of recovery. Using a segmentation approach helps control spend and reduces labor costs to maximize the dollars collected. Credit based data can be utilized in decision trees to create segments that can be used with or without collection models. For example, below is a portion of a full decision tree that shows the separation in the liquidation rates by applying an attribute to a recovery score This entire segment has an average of 21.91 percent liquidation rate. The attribute applied to this score segment is the aggregated available credit on open bank card trades updated within 12 months. By using just this one attribute for this score band, we can see that the liquidation rates range from 11 to 35 percent. Additional attributes can be applied to grow the tree to isolate additional pockets of customers that are more recoverable, and identify segments that are not likely to be recovered. From a fully-developed segmentation analysis, appropriate collections strategies can be determined to prioritize those accounts that are most likely to pay, creating new efficiencies within existing collection strategies to help improve collections.

By: Roger Ahern It’s been proven in practice many times that by optimizing decisions (through improved decisioning strategies, credit risk modeling, risk-based pricing, enhanced scoring models, etc.) you will realize significant business benefits in key metrics, such as net interest margin, collections efficiency, fraud referral rates and many more.  However, given that a typical company may make more than eight million decisions per year, which decisions should one focus on to deliver the greatest business benefit? In working with our clients, Experian has compiled the following list of relevant types of decisions that can be improved through improvements in decision analytics.  As you review the list below, you should identify those decisions that are relevant to your organization, and then determine which decision types would warrant the greatest opportunity for improvement. • Cross-sell determination • Prospect determination • Prescreen decision • Offer/treatment determination • Fraud determination • Approve/decline decision • Initial credit line/limit/usage amount • Initial pricing determination • Risk-based pricing • NSF pay/no-pay decision • Over-limit/shadow limit authorization • Credit line/limit/usage/ management • Retention decisions • Loan/payment modification • Repricing determination • Predelinquency treatment • Early/late-stage delinquency treatment • Collections agency placement • Collection/recovery treatment  

In my previous two blogs, I introduced the definition of strategic default and compared and contrasted the population to other types of consumers with mortgage delinquency.  I also reviewed a few key characteristics that distinguish strategic defaulters as a distinct population. Although I’ve mentioned that segmenting this group is important, I would like to specifically discuss the value of segmentation as it applies to loan modification programs and the selection of candidates for modification. How should loan modification strategies be differentiated based on this population? By definition, strategic defaulters are more likely to take advantage of loan modification programs. They are committed to making the most personally-lucrative financial decisions, so the opportunity to have their loan modified - extending their ‘free’ occupancy – can be highly appealing.  Given the adverse selection issue at play with these consumers, lenders need to design loan modification programs that limit abuse and essentially screen-out strategic defaulters from the population. The objective of lenders when creating loan modification programs should be to identify consumers who show the characteristics of cash-flow managers within our study. These consumers often show similar signs of distress as the strategic defaulters, but differentiate themselves by exhibiting a willingness to pay that the strategic defaulter, by definition, does not. So, how can a lender make this identification? Although these groups share similar characteristics at times, it is recommended that lenders reconsider their loan modification decisioning algorithms, and modify their loan modification offers to screen out strategic defaulters.  In fact, they could even develop programs such as equity-sharing arrangements whereby the strategic defaulter could be persuaded to remain committed to the mortgage.  In the end, strategic defaulters will not self-identify by showing lower credit score trends, by being a bank credit risk, or having previous bankruptcy scores, so lenders must create processes to identify them among their peers. For more detailed analyses, lenders could also extend the Experian-Oliver Wyman study further, and integrate additional attributes such as current LTV, product type, etc. to expand their segment and identify strategic defaulters within their individual portfolios.    

--by Andrew Gulledge General configuration issues Question selection- In addition to choosing questions that generally have a high percentage correct and fraud separation, consider any questions that would clearly not be a fit to your consumer population. Don’t get too trigger-happy, however, or you’ll have a spike in your “failure to generate questions” rate. Number of questions- Many people use three or four out-of-wallet questions in a Knowledge Based Authentication session, but some use more or less than that, based on their business needs. In general, more questions will provide a stricter authentication session, but might detract from the customer experience. They may also create longer handling times in a call center environment. Furthermore, it is harder to generate a lot of questions for some consumers, including thin-file types. Fewer Knowledge Based Authentication questions can be less invasive for the consumer, but limits the fraud detection value of the KBA process. Multiple choice- One advantage of this answer format is that it relies on recognition memory rather than recall memory, which is easier for the consumer. Another advantage is that it generally prevents complications associated with minor numerical errors, typos, date formatting errors and text scrubbing requirements. A disadvantage of multiple-choice, however, is that it can make educated guessing (and potentially gaming) easier for fraudsters. Fill in the blank- This is a good fit for some KBA questions, but less so with others. A simple numeric answer works well with fill in the blank (some small variance can be allowed where appropriate), but longer text strings can present complications. While undoubtedly difficult for a fraudster to guess, for example, most consumers would not know the full, official and (correct spelling) of the name to which they pay their monthly auto payment. Numeric fill in the blank questions are also good candidates for KBA in an IVR environment, where consumers can use their phone’s keypad to enter the answers.  

A recent New York Times (1) article outlined the latest release of credit borrowing by the Federal Reserve, indicating that American’s borrowed less for the ninth-straight month in October. Nested within the statistics released by the Federal Reserve were metrics around reduced revolving credit demand and comments about how “Americans are borrowing less as they try to replenish depleted investments.” While this may be true, I tend to believe that macro-level statements are not fully explaining the differences between consumer experiences that influence relationship management choices in the current economic environment. To expand on this, I think a closer look at consumers at opposite ends of the credit risk spectrum tells a very interesting story. In fact, recent bank card usage and delinquency data suggests that there are at least a couple of distinct patterns within the overall trend of reducing revolving credit demand: • First, although it is true that overall revolving credit balances are decreasing, this is a macro-level trend that is not consistent with the detail we see at the consumer level. In fact, despite a reduction of open credit card accounts and overall industry balances, at the consumer-level, individual balances are up – that’s to say that although there are fewer cards out there, those that do have them are carrying higher balances. • Secondly, there are significant differences between the most and least-risky consumers when it comes to changes in balances. For instance, consumers who fall into the least-risky VantageScore® tiers, Tier A and B, show only 12 percent and 4 percent year-over-year balance increases in Q3 2009, respectively. Contrast that to the increase in average balance for VantageScore F consumers, who are the most risky, whose average balances increased more than 28 percent for the same time period. So, although the industry-level trend holds true, the challenges facing the “average” consumer in America are not average at all – they are unique and specific to each consumer and continue to illustrate the challenge in assessing consumers' credit card risk in the current credit environment. 1 http://www.nytimes.com/2009/12/08/business/economy/08econ.html  

In my last blog, I discussed the presence of strategic defaulters and outlined the definitions used to identify these consumers, as well as other pools of consumers within the mortgage population that are currently showing some measure of mortgage repayment distress. In this section, I will focus on the characteristics of strategic defaulters, drilling deeper into the details behind the population and learning how one might begin to recognize them within that population. What characteristics differentiate strategic defaulters? Early in the mortgage delinquency stage, mortgage defaulters and cash flow managers look quite similar – both are delinquent on their mortgage, but are not going bad on any other trades. Despite their similarities, it is important to segment these groups, since mortgage defaulters are far more likely to charge-off and far less likely to cure than cash flow managers. So, given the need to distinguish between these two segments, here are a few key measures that can be used to define each population. Origination VantageScore® credit score • Despite lower overall default rates, prime and super-prime consumers are more likely to be strategic defaulters  Origination Mortgage Balance • Consumers with higher mortgage balances at origination are more likely to be strategic defaulters, we conclude this is a result of being further underwater on their real estate property than lower-balance consumers Number of Mortgages • Consumers with multiple first mortgages show higher incidence of strategic default.  This trend represents consumers with investment properties making strategic repayment decisions on investments (although the majority of defaults still occur on first mortgages where the consumer has only one first mortgage) Home Equity Line Performance • Strategic defaulters are more likely to remain current on Home Equity Lines until mortgage delinquency occurs, potentially a result of drawing down the HELOC line as much as possible before becoming delinquent on the mortgage Clearly, there are several attributes that identify strategic defaulters and can assist in differentiating them from cash flow managers. The ability to distinguish between these two populations is extremely valuable when considering its usefulness in the application of account management and collections management, improving collections, and loan modification, which is my next topic. Source: Experian-Oliver Wyman Market Intelligence Reports; Understanding strategic default in mortgage topical study/webinar, August 2009.

I have already commented on “secret questions” as the root of all evil when considering tools to reduce identity theft and minimize fraud losses.  No, I’m not quite ready to jump off  that soapbox….not just yet, not when we’re deep into the season of holiday deals, steals and fraud.  The answers to secret questions are easily guessed, easily researched, or easily forgotten.  Is this the kind of security you want standing between your account and a fraudster during the busiest shopping time of the year? There is plenty of research demonstrating that fraud rates spike during the holiday season.  There is also plenty of research to demonstrate that fraudsters perpetrate account takeover by changing the pin, address, or e-mail address of an account – activities that could be considered risky behavior in decisioning strategies.  So, what is the best approach to identity theft red flags and fraud account management?  A risk based authentication approach, of course! Knowledge Based Authentication (KBA) provides strong authentication and can be a part of a multifactor authentication environment without a negative impact on the consumer experience, if the purpose is explained to the consumer.  Let’s say a fraudster is trying to change the pin or e-mail address of an account.  When one of these risky behaviors is initiated, a Knowledge Based Authentication session begins. To help minimize fraud, the action is prevented if the KBA session is failed.  Using this same logic, it is possible to apply a risk based authentication approach to overall account management at many points of the lifecycle: • Account funding • Account information change (pin, e-mail, address, etc.) • Transfers or wires • Requests for line/limit increase • Payments • Unusual account activity • Authentication before engaging with a fraud alert representative Depending on the risk management strategy, additional methods may be combined with KBA; such as IVR or out-of-band authentication, and follow-up contact via e-mail, telephone or postal mail.  Of course, all of this ties in with what we would consider to be a comprehensive Red Flag Rules program. Risk based authentication, as part of a fraud account management strategy, is one of the best ways we know to ensure that customers aren’t left singing, “On the first day of Christmas, the fraudster stole from me…”  

--by Andrew Gulledge Where does Knowledge Based Authentication fit into my decisioning strategy? Knowledge Based Authentication can fit into various parts of your authentication process. Some folks choose to put every consumer through KBA, while others only send their riskier transactions through the out-of-wallet questions. Some people use Knowledge Based Authentication to feed a manual review process, while others use a KBA failure as a hard-decline. Uses for KBA are as sundry and varied as the questions themselves. Decision Matrix- As discussed by prior bloggers, a well-engineered fraud score can provide considerable lift to any fraud risk strategy. When possible, it is a good idea to combine both score and questions into the decisioning process. This can be done with a matrixed approach—where you are more lenient on the questions if the applicant has a good fraud score, and more lenient on the score if the applicant did well on the questions. In a decision matrix, a set decision code is placed within various cells, based on fraud risk. Decision Overrides- These provide a nice complement to your standard fraud decisioning strategy. Different fraud solution vendors provide different indicators or flags with which decisioning rules can be created. For example, you might decide to fail a consumer who provides a social security number that is recorded as deceased. These rules can help to provide additional lift to the standard decisioning strategy, whether it is in addition to Knowledge Based Authentication questions alone, questions and score, etc. The overrides can be along the lines of both auto-pass and auto-fail.  

By: Wendy Greenawalt In my last blog on optimization we discussed how optimized strategies can improve collection strategies. In this blog, I would like to discuss how optimization can bring value to decisions related to mortgage delinquency/modification. Over the last few years mortgage lenders have seen a sharp increase in the number of mortgage account delinquencies and a dramatic change in consumer mortgage payment trends.   Specifically, lenders have seen a shift in consumer willingness from paying their mortgage obligation first, while allowing other debts to go delinquent. This shift in borrower behavior appears unlikely to change anytime soon, and therefore lenders must make smarter account management decisions for mortgage accounts. Adding to this issue, property values continue to decline in many areas and lenders must now identify if a consumer is a strategic defaulter, a candidate for loan modification, or a consumer affected by the economic downturn. Many loans that were modified at the beginning of the mortgage crisis have since become delinquent and have ultimately been foreclosed upon by the lender. Making optimizing decisions related to collection action for mortgage accounts is increasingly complex, but optimization can assist lenders in identifying the ideal consumer collection treatment. This is taking place while lenders considering organizational goals, such as minimizing losses and maximizing internal resources, are retaining the most valuable consumers. Optimizing decisions can assist with these difficult decisions by utilizing a mathematical algorithm that can assess all possible options available and select the ideal consumer decision based on organizational goals and constraints. This technology can be implemented into current optimizing decisioning processes, whether it is in real time or batch processing, and can provide substantial lift in prediction over business as usual techniques.    

For the past couple years, the deterioration of the real estate market and the economy as a whole has been widely reported as a national and international crisis. There are several significant events that have contributed to this situation, such as, 401k plans have fallen, homeowners have simply abandoned their now under-valued properties, and the federal government has raced to save the banking and automotive sectors. While the perspective of most is that this is a national decline, this is clearly a situation where the real story is in the details. A closer look reveals that while there are places that have experienced serious real estate and employment issues (California, Florida, Michigan, etc.), there are also areas (Texas) that did not experience the same deterioration in the same manner. Flash forward to November, 2009 – with signs of recovery seemingly beginning to appear on the horizon – there appears to be a great deal of variability between areas that seem poised for recovery and those that are continuing down the slope of decline. Interestingly though, this time the list of usual suspects is changing. In a recent article posted to CNN.com, Julianne Pepitone observes that many cities that were tops in foreclosure a year ago have since shown stabilization, while at the same time, other cities have regressed. A related article outlines a growing list of cities that, not long ago, considered themselves immune from the problems being experienced in other parts of the country. Previous economic success stories are now being identified as economic laggards and experiencing the same pains, but only a year or two later. So – is there a lesson to be taken from this? From a business intelligence perspective, the lesson is generalized reporting information and forecasting capabilities are not going to be successful in managing risk. Risk management and forecasting techniques will need to be developed around specific macro- and micro-economic changes.  They will also need to incorporate a number of economic scenarios to properly reflect the range of possible future outcomes about risk management and risk management solutions. Moving forward, it will be vital to understand the differences in unemployment between Dallas and Houston and between regions that rely on automotive manufacturing and those with hi-tech jobs. These differences will directly impact the performance of lenders’ specific footprints, as this year’s “Best Place to Live” according to Money.CNN.com can quickly become next year’s foreclosure capital. ihttp://money.cnn.com/2009/10/28/real_estate/foreclosures_worst_cities/index.htm?postversion=2009102811 iihttp://money.cnn.com/galleries/2009/real_estate/0910/gallery.foreclosures_worst_cities/2.html  

By: Wendy Greenawalt Optimization has become a "buzz word" in the financial services marketplace, but some organizations still fail to realize all the possible business applications for optimization. As credit card lenders scramble to comply with the pending credit card legislation, optimization can be a quick and easily implemented solution that fits into current processes to ensure compliance with the new regulations. Optimizing decisions Specifically, lenders will now be under strict guidelines of when an APR can be changed on an existing account, and the specific circumstances under which the account must return to the original terms. Optimization can easily handle these constraints and identify which accounts should be modified based on historical account information and existing organizational policies. APR account changes can require a great deal of internal resources to implement and monitor for on-going performance. Implementing an optimized strategy tree within an existing account management strategy will allow an organization to easily identify consumer level decisions.  This can be accomplished while monitoring accounts through on-going batch processing. New delivery options are now available for lenders to receive optimized strategies for decisions related to: Account acquisition Customer management Collections Organizations who are not currently utilizing this technology within their  processes should investigate the new delivery options. Recent research suggests optimizing decisions can provide an improvement of 7-to-16 percent over current processes.  

In my last blog, I discussed the basic concept of a maturation curve, as illustrated below: Exhibit 1 In Exhibit 1, we examine different vintages beginning with those loans originated by year during Q2 2002 through Q2 2008. The purpose of the vintage analysis is to identify those vintages that have a steeper slope towards delinquency, which is also known as delinquency maturation curve. The X-axis represents a timeline in months, from month of origination.  Furthermore, the Y-axis represents the 90+ delinquency rate expressed as a percentage of balances in the portfolio.  Those vintage analyses that have a steeper slope have reached a normalized level of delinquency sooner, and could in fact, have a trend line suggesting that they overshoot the expected delinquency rate for the portfolio based upon credit quality standards. So how can you use a maturation curve as a useful portfolio management tool? As a consultant, I spend a lot of time with clients trying to understand issues, such as why their charge-offs are higher than plan (budget).  I also investigate whether the reason for the excess credit costs are related to collections effectiveness, collections strategy, collections efficiency, credit quality or a poorly conceived budget. I recall one such engagement, where different functional teams within the client’s organization were pointing fingers at each other because their budget evaporated. One look at their maturation curves and I had the answers I needed. I noticed that two vintages per year had maturation curves that were pointed due north, with a much steeper curve than all other months of the year. Why would only two months or vintages of originations each year be so different than all other vintage analyses in terms of performance? I went back to my career experiences in banking, where I worked for a large regional bank that ran marketing solicitations several times yearly. Each of these programs was targeted to prospects that, in most instances, were out-of-market, or in other words, outside of the bank’s branch footprint. Bingo! I got it! The client was soliciting new customers out of his market, and was likely getting adverse selection. While he targeted the “right” customers – those with credit scores and credit attributes within an acceptable range, the best of that targeted group was not interested in accepting their offer, because they did not do business with my client, and would prefer to do business with an in-market player. Meanwhile, the lower grade prospects were accepting the offers, because it was a better deal than they could get in-market. The result was adverse selection...and what I was staring at was the "smoking gun" I’d been looking for with these two-a-year vintages (vintage analysis) that reached the moon in terms of delinquency. That’s the value of building a maturation curve analysis – to identify specific vintages that have characteristics that are more adverse than others.  I also use the information to target those adverse populations and track the performance of specific treatment strategies aimed at containing losses on those segments. You might use this to identify which originations vintages of your home equity portfolio are most likely to migrate to higher levels of delinquency; then use credit bureau attributes to identify specific borrowers for an early lifecycle treatment strategy. As that beer commercial says – “brilliant!”  

--by Jeff Bernstein In the current economic environment, many lenders and issuers across the globe are struggling to manage the volume of caseloads coming into collections. The challenge is that as these new collection cases come into collections in early phases of delinquency, the borrower is already in distress, and the opportunity to have a good outcome is diminished. One of the real “hot” items on the list of emerging best practices and innovating changes in collections is the concept of early lifecycle treatment strategy. Essentially, what we are referring to is the treatment of current and non-delinquent borrowers who are exhibiting higher risk characteristics.  There are also those who are at-risk of future default at higher levels than average. The challenge is how to identify these customers for early intervention and triage in the collections strategy process. One often-overlooked tool is the use of maturation curves to identify vintages within a portfolio that is performing worse than average. A maturation curve identifies how long from origination until a vintage or segment of the portfolio reaches a normalized rate of delinquency. Let’s assume that you are launching a new credit product into the marketplace. You begin to book new loans under the program in the current month. Beyond that month, you monitor all new loans that were originated/booked during that initial time frame which we can identify as a “vintage” of the portfolio. Each month’s originations are a separate vintage or vintage analysis, and we can track the performance of each vintage over time. How many months will it take before the “portfolio” of loans booked in that initial month reach a normal level of delinquency based on these criteria: the credit quality of the portfolio and its borrowers, typical collections servicing, delinquency reporting standards, and factor of time?  The answer would certainly depend upon the aforementioned factors, and could be graphed as follows:   Exhibit 1        In Exhibit 1, we examine different vintages beginning with those loans originated during Q2 2002, and by year Q2 2008. The purpose of the analysis is to identify those vintages that have a steeper slope towards delinquency, which is also known as a delinquency maturation curve.  The X-axis represents a timeline in months, from month of origination.  Furthermore,, the Y-axis represents the 90+ delinquency rate expressed as a percentage of balances in the portfolio. Those vintages that have a steeper slope have reached a normalized level of delinquency sooner, and could in fact, have a trend line suggesting that they overshoot the expected delinquency rate for the portfolio based upon credit quality standards. So how do we use the maturation curve as a tool? In my next blog, I will discuss how to use maturation curves to identify trends across various portfolios.  I will also examine differentiate collections issues from originations or lifecycle risk management opportunities.    

In my last post I discussed the problem with confusing what I would call “real” Knowledge Based Authentication (KBA) with secret questions.   However, I don’t think that’s where the market focus should be.  Instead of looking at Knowledge Based Authentication (KBA) today, we should be looking toward the future, and the future starts with risk-based authentication. If you’re like most people, right about now you are wondering exactly what I mean by risk-based authentication.  How does it differ from Knowledge Based Authentication, and how we got from point A to point B? It is actually pretty simple.  Knowledge Based Authentication is one factor of a risk-based authentication fraud prevention strategy.  A risk- based authentication approach doesn’t rely on question/answers alone, but instead utilizes fraud models that include Knowledge Based Authentication performance as part of the fraud analytics to improve fraud detection performance.  With a risk-based authentication approach, decisioning strategies are more robust and should include many factors, including the results from scoring models. That isn’t to say that Knowledge Based Authentication isn’t an important part of a risk-based approach.  It is.  Knowledge Based Authentication is a necessity because it has gained consumer acceptance. Without some form of Knowledge Based Authentication, consumers question an organization’s commitment to security and data protection. Most importantly, consumers now view Knowledge Based Authentication as a tool for their protection; it has become a bellwether to consumers. As the bellwether, Knowledge Based Authentication has been the perfect vehicle to introduce new and more complex authentication methods to consumers, without them even knowing it.  KBA has allowed us to familiarize consumers with out-of-band authentication and IVR, and I have little doubt that it will be one of the tools to play a part in the introduction of voice biometrics to help prevent consumer fraud. Is it always appropriate to present questions to every consumer?  No, but that’s where a true risk-based approach comes into play.  Is Knowledge Based Authentication always a valuable component of a risk based authentication tool to minimize fraud losses as part of an overall approach to fraud best practices?  Absolutely; always. DING!