When asked what made him an outstanding hockey player Wayne Gretzky replied,” I skate where the puck is going, not where it has been”. Unlike today’s credit characteristics which are based upon an historical perspective of a consumer’s behavior, this new breed of credit characteristics anticipate future consumer credit profiles allowing lenders to tailor credit offerings and risk management strategies based upon where the consumer is headed, rather than where they have been.
Traditional credit characteristics and scoring systems rely exclusively on the consumer’s historical credit profile, which is analogous to driving a car while only looking through the rearview mirror. Today’s decision support and credit scoring systems “reward” and “penalize” consumers based upon observed credit behavior, ignoring obvious and imminent credit profile changes on the horizon. For instance, consumers are penalized for derogatory credit performance until it is removed from a consumer’s credit file. Typically more recent derogatory information is treated more severely and is gradually “discounted” until the information is removed from the credit file, As a result, given all other things equal, as derogatory information ages a consumer’s credit risk score gradually improves. When derogatory information eventually drops of the credit file due to regulatory requirements the consumer’s credit risk score normally increase significantly. The timing as to when derogatory information drops from a consumer’s credit file is predictable, why not take advantage of this event in acquisition and retention programs before it happens? In addition to derogatory information dropping of the file, there are several other predictable changes to neutral or positive credit characteristics; such as age of oldest tradeline, number of inquiries within a specific period of time and age of most recent tradeline that will generally increase a consumer’s credit score once they reach a certain age. Because the relationship between the timing and age of specific credit events and risk level is well understood, it is surprising that forward looking credit characteristics and scores for specific credit programs (i.e. acquisition, retention, line increase) have not been developed and deployed.
Forward Looking Solutions May Offer Better Value that Reactive Trigger Solutions
Every night at the stroke of midnight, tens of thousands U.S. consumers experience a change to their credit file and each of the “Big Three” U.S. credit reporting agencies pass these credit profiles through sophisticated monitoring or “triggering” systems which capture, record, sort, rank and notify lenders about consumers and account holders with meaningful changes.. These trigger systems are an highly effective approach to alert lenders about a variety of credit changes that are proven, leading indicators of future consumer behavior. One reason why these systems are so effective is that the changes identified are recent, enabling lenders to take appropriate action. Although trigger systems are highly effective, they reaction to changes to a consumer’s credit profile that have taken place and are not truly forward looking because they do not incorporate imminent credit changes that will occur at a specific point in time. The expectation is that forward looking characteristics and scores, which anticipate imminent changes to a consumer’s credit profile, may more accurately reflect the consumer’s risk profile. By identifying these consumers before their credit score improves a lender may the gain a competitive advantage by offering a credit product or line increase ahead of their competitors. Unfortunately champion/challenger test data on this concept does not exist.
Modifications of Existing Analytic Platforms May Accelerate Forward Looking Solutions
The infrastructure necessary to test the performance of forward looking characteristics and scores most likely exists within credit bureaus that currently provide credit characteristics and scores. Modifications to existing credit bureau based characteristic and scoring systems to create forward looking systems would be fairly straightforward requiring; 1) the ability to identify and exclude delinquent tradelines, collection accounts, public records and inquires that will drop off consumer credit files at a specific point in time (i.e. three months from the current date), 2) a process to determine the age of tradelines, collections accounts public records, and inquiries that will remain on a consumer’s credit file at a specific point in time, and 3) a process to calculate and return credit characteristics and scores based upon anticipated consumer credit files. The underlying assumption for information associated with the remaining tradelines, collection accounts and public records is that the delinquency level, current balance and credit limit will not change.
Forward Looking Solutions Requires Forward Looking Leadership
Forward looking solutions may also be greatly enhanced by with the ability to incorporate projections of future levels of consumer debt and credit utilization based upon the trajectory of time series credit balance, limit and payment information recently made available at U.S. credit bureaus. Modifications to existing credit characteristic and scoring platforms that incorporate projected debt levels and credit utilization may entail significantly more resources but may have a broader impact, especially from a risk profile perspective. Given the potentially significant amount of resources required to make create forward looking solutions, commitment from either forward looking leadership within a credit bureau or a lender is required to modify existing credit characteristics and scoring systems and to test this new concept. It will be interesting to see which party, if any, takes the initiative to pursue this concept.
About the Author: Chet Wiermanski is one of BIIA’s contributing editors writing on the subjects of credit scoring and decision systems. He is a Visiting Scholar at the Federal Reserve Bank of Philadelphia researching new applications of consumer credit report information. Additionally, Chet is Managing Director of Aether Analytics which specializes on leveraging hidden data sequences and time series components within consumer credit information typically ignored by traditional credit bureau based solutions. Previously Chet was the Global Chief Scientist at TransUnion LLC. Holding a variety of positions within TransUnion, during his tenure, between July 1997 and February 2012, he was responsible for identifying, evaluating and developing new technology platforms involving alternative data sources, predictive modeling, econometric forecasting and related consulting services.