Score Complete
Customer Handbook
Prepared by:
Larry Macdonald, Sr. Product Manager
10-Jun-2014
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 2
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Table of Contents
Introducing Score Complete ........................................................................................................................................... 3
1. Modeling Concepts ................................................................................................................................................ 3
1.1 Population of Interest / Minimum Scoring Criteria ........................................................................................ 4
1.2 Response Variable ....................................................................................................................................... 5
1.3 Available Data / Independent Variables ........................................................................................................ 6
2. The Score Complete Model ................................................................................................................................... 7
2.1 Segmentation ............................................................................................................................................... 7
2.2 Modeling Technique ..................................................................................................................................... 8
2.3 Scaling .......................................................................................................................................................... 9
2.4 Attributes .................................................................................................................................................... 11
2.5 Account Management and Scorecard Migration ......................................................................................... 12
3. Output ................................................................................................................................................................. 14
4. Using Score Complete – Alone or With Another Score ....................................................................................... 15
5. Evaluating a Model .............................................................................................................................................. 15
5.1 Score Complete Development Data ........................................................................................................... 18
Summary ...................................................................................................................................................................... 20
Additional Reading ....................................................................................................................................................... 20
Appendix – Reason Codes, Scorecard Indicator, Reject Codes .................................................................................. 21
Reason Codes ......................................................................................................................................................... 21
Reject Codes ........................................................................................................................................................... 23
Score Card Indicators .............................................................................................................................................. 23
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 3
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Introducing Score Complete
Risk managers have a difficult task. Their companies are in the business of lending money or granting credit to
ordinary Canadians who need credit cards for everyday purchases, loans to buy goods and services, lines of credit to
optimize and bring flexibility to credit management, and mortgages so they will have a place to live. Unfortunately, it
is not profitable for these companies if their customers don’t pay them back. A credit file, supplemented with a credit
score, gives a risk manager the ability to assess the likelihood that a customer will meet their financial obligations – to
make regular payments on the credit that they use.
Credit scores are systematic and predictive, enabling the application of consistent business rules. Low risk
customers may receive better product offerings, better terms, or higher limits, while high risk customers may be
required to provide securitization or may not be offered credit at all.
Most Canadian consumers regularly pay their bills and have established good credit histories. These are profitable
consumers for the lending institutions in the financial industry. Unfortunately, the consumers who are non-payers
cause significant losses for the lenders, resulting in increased interest rates for all. Identifying these consumers and
mitigating their losses in a timely manner makes the entire lending practice more efficient for the lender and borrower
alike.
Score Complete is our most accurate solution in the prediction of consumer delinquency risk. Consumers or
applicants with low scores have a high probability of going 90 days past due or worse on their debt obligations over
the next 12 months. But Score Complete attempts to do something that no other score in the market does. It tries to
score every Canadian consumer file or application for credit.
Score Complete uses such credit file characteristics as missed payments, utilization and balances, inquiries, public
records, and the ages and types of credit products, to assess delinquency risk. For files with limited credit
information and for applicants without credit files, aggregated credit data from the consumer’s neighbourhood is used
to augment the limited information available to assess credit risk.
1. Modeling Concepts
It is a relatively simple task to determine a consumer’s behaviour when there is a long history with a lot of accounts
and activity. One can assume that prior patterns will repeat. Someone who has paid their bills regularly for a number
of years may be expected to continue to do so. When there is less information available, other considerations must
be taken into account. Deriving the relationship between the information in a person’s credit file and the future
behaviour is a statistical exercise, creating the proper weighting for each factor, by determining the characteristics
that distinguish the good payers from those who go delinquent. These relationships among the various factors may
shift over time, due to different market conditions, so using a score that is up to date is key.
Score Complete is a predictive model offered by Equifax that risk managers use to help them determine which
customers or applicants are creditworthy; that have credit characteristics that are associated with good payers.
In order to build a model, three key components are required: population, outcome, and data. The population is the
collection of records used to build the model, and should be representative of the population where the model will be
used. The outcome is the value to be modelled, and represents the unknown quantity that the model predicts. The
data are the known attributes that are available at the time and used in the calculations.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 4
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
1.1 Population of Interest / Minimum Scoring Criteria
There are two considerations necessary to define a population for a credit model. The first is to decide on the target
population where the model is of interest. For a delinquency model, the ideal population may be any Canadian
consumer or applicant for credit. Assessing the risk of every prospect or existing customer is the best case.
The second consideration is available information. It may be difficult or impossible to make an accurate assessment
of risk given the sparcity of data in some cases. Among the common limitations:
Credit File Not Found. Sometimes an applicant does not have a credit file. This includes young
Canadians who apply for credit for the first time, or new immigrants. Some people may go for many years
without building a credit history, preferring to pay cash for everything so as not to owe any money, or they
use their family’s credit (parents or spouse) for their own needs. A change in their family status, such as
death or divorce, forces them to establish their own credit. Additionally, a credit file may not be found in
cases where there are mismatches between the information in the credit file and the application. This could
be due to typographical errors, information variants such as different name versions like Robert and Bob,
revised information that hasn’t been updated on the credit file like new address or change of name, or
format errors, such as having the input data in the wrong fields, or supplying an address out of the country.
Death Notice on File. Although the estate of a deceased individual may be responsible for the financial
commitments in some cases, the information in the credit file may not be predictive of the future
performance of the account.
Inactivity. If a credit file has not been updated for a period of time, the information that it contains may be
stale. In many cases, it may be more accurate for an institution to make lending decisions based on other
information, such as income statements or wealth (including property equity), by providing collateral or
security, or having a co-signer.
The combination of the target population and the available information to generate a score is known as the minimum
scoring criteria.
For Score Complete, consideration was given to whether the credit file, by itself, contained enough information to
calculate the credit score. It was determined that the criteria for this was to include all Canadian consumers with
credit activity within the last 24 months and at least one valid trade line
1
. Activity is defined as a trade line updated
(based on the date reported) or a hard inquiry
2
.
For consumers with credit files that did not meet these criteria, aggregated credit data was used to augment the
amount of available information about the consumer. If either a credit file is found, or aggregated credit data is found,
Score Complete estimates the risk and calculates a score. Score Complete calculates and returns a score even
when a credit file is not found.
1
Trade lines are the accounts or credit products that financial institutions report to the credit bureau. They are the credit cards,
loans, lines of credit, mortgages, etc. that show what responsibilities and history the consumers have with various reporting
institutions.
2
Inquiries are posted whenever somebody views or receives information contained in the credit report. Hard inquiries indicate
that a consumer has applied for credit and granted permission for someone to see their credit report for the purpose of
adjudicating that credit. Soft inquiries are posted whenever a company refreshes information about their customers, but are not
motivated by consumer activity. Soft inquiries are only visible to the consumer, and do not affect any credit scores.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 5
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
For credit files with death notices, the minimum possible score of 300 is returned. Many algorithms default to a no
score result with a death notice. Since a no score implies that the institution must investigate further in order to
assess the applicant, Score Complete suggests that no further investigation is required.
There are only two cases where Score Complete will not return a valid score. If there is no credit file, and there is no
aggregated credit data from the address, there is no information available to the score. The other case is if a file has
been flagged by Equifax as a manual file, or file under review. Manual files are rare; almost all instances of failure for
Score Complete will be due to problems with the input address.
1.2 Response Variable
Credit scoring is done to help risk managers understand how likely it is that their customers are going to make the
required payments on their credit products (loans, credit cards, lines of credit, mortgages, etc.). The model requires
taking a representative sample from the target population from a recent archive period (known as the observation
point), and then defining and calculating the response variable; consumer credit files are observed at a more recent
period (12 months after the observation period for Score
Complete; this is called the performance window) to
determine whether there has been negative behaviour.
For Score Complete, this response variable is defined as
a serious delinquency (90 days or worse) or write-off of a
trade line, or presence of a derogatory public record
such as a bankruptcy, within the performance window.
For the development of Score Complete, three different
time intervals were used to account for seasonality.
Approximately 2 million records without death notices
were chosen from time periods (observation and
performance) spaced 12 months apart:
June 2010 and June 2011
September 2010 and September 2011
December 2010 and December 2011
In addition, approximately 900,000 files were used for consumers who opened a new account between July 2010 and
March 2011, and did not meet the criteria to score based solely on the credit file, in order to build the additional
scorecards.
Special emphasis was placed on detecting characteristics of consumers at origination who opened accounts that
went delinquent in the first year. Since Score Complete is particularly strong at predicting this behavior, it is an
excellent score for risk managers to use at perhaps the most critical point of the consumer’s life cycle – the point at
which the consumer becomes a customer of the risk manager’s lending institution.
While Score Complete was developed as a score that predicts 90-day or worse delinquency of a consumer of any
trade within 12 months, it is also very predictive of other similar outcomes, such as delinquency within 24 months
rather than 12, or 60-day delinquencies rather than 90-days. Score Complete can be used to predict the likelihood of
Special emphasis was placed on detecting
characteristics of consumers at origination who
opened accounts that went delinquent in the first
year. Since Score Complete is particularly strong
at predicting this behavior, it is an excellent score
for risk managers to use at perhaps the most
critical point of the consumer’s life cycle – the
point at which the consumer becomes a
customer of the risk manager’s lending
institution.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 6
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
a consumer going delinquent on any individual account when being assessed for account management purposes, as
well as predicting delinquency for a new account during acquisition / adjudication.
1.3 Available Data / Independent Variables
A model estimates an unknown quantity by developing the relationship between the known attributes and the
required outcome (the performance as defined in the previous section). The known attributes have to be available at
the time that the unknown quantity is to be estimated. The statistical term for these attributes is independent
variables. For Score Complete, the attributes come from the credit file in most cases. When the credit file is not
robust enough to return a score reliably, additional attributes are available from the aggregated data available within
Neighbourhood View
3
.
The credit file attributes are the Equifax Canada Risk Modelling Segments (RMS), consisting of over 400 proprietary
credit file attributes covering a wide spectrum of credit file characteristics including delinquency, utilization and
balances, inquiries, public records, and the make-up of the wallet. These segments include many industry-specific
attributes as well as some that are aggregated for all trades or inquiries.
In order for Score Complete to accurately predict a delinquency rate for consumers with very limited credit file
information, or no credit file at all, an additional data source must be used. Using the information that is there along
with additional information that is relevant to the consumer and predictive of his behavior allows Score Complete to
be a complete score for the Canadian credit-seeking population.
Augmenting the credit file in cases where there may not be enough information to calculate a reliable score is
aggregated credit data from Neighbourhood View. Neighbourhood View is a tool originally designed for marketers
who are looking for consumers that have desirable credit characteristics and are more likely to have a desire to
consumer the company’s products. Neighbourhood View is aggregated credit data, where the credit histories of
consumers in each neighbourhood are combined to give a profile.
Information aggregated to a neighbourhood is representative, in a majority of cases, of the individuals within the
neighbourhood. The average credit profile of a neighbourhood would be consistent with the credit profile of most of
the residents in that neighbourhood. This consistency is enhanced by the fact that the consumer and all members of
the same household contribute data to the neighbourhood. Therefore, it is reasonable that the aggregate information
of the neighbourhood makes a good proxy for individual data, when individual data cannot be obtained.
The neighbourhood in which a person lives may be a defining characteristic. Whether they live in a rich or poor
neighbourhood, a house, apartment, or condominium, a rural or urban community, these characteristics are shared
with their closest neighbours. Knowing whether the other residents in a neighbourhood pay their bills on time or not,
carry high or balances on their accounts, and whether or not they use different credit products reflects on the
individuals for which little information is known. This aggregated data is helpful in predicting whether a consumer is
likely to pay regularly their own bills on time.
With Score Complete, neighbourhoods are geographically defined at the Street Level. While privacy legislation
prevents aggregating fewer than 15 credit files together, postal codes are sometimes very large, and one may wish to
3
When files have a lot of information, predicting behavior is easy. With less information, a good model can provide an excellent
estimate. When the credit file has very little information, other sources are required in order to build a model that distinguishes
the consumers who are good payers from those who will go delinquent.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 7
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
split a large postal code into smaller segments. This is what Street Level does. The credit files in each postal code
are sorted by street, house (civic) number, and apartment (or suite) number. Along each street, credit files are
counted by address until at least 15 credit files are found. Additional credit files may be included if they match the
address of the 15
th
file. This segment is called a subdivision. Then the process continues from the next address. In
this manner, large postal codes are segmented into a number of smaller subdivisions with at least 15 credit files, so
these subdivisions are much more granular; perhaps only 7 or 8 households are used to define them. Consumers
are aggregated with fewer neighbours, and likely those who are in closer proximity, and therefore likely share more
common characteristics than those in a larger area. In addition, since fewer consumers are aggregated together, an
individual’s information contributes a larger proportion of data to the subdivision, making the subdivision data a much
better proxy for the individual than the postal code as a whole.
2. The Score Complete Model
The Score Complete model returns a three-digit numerical score that corresponds to the delinquency risk for the
individual consumer with the given credit file and address information. Consumers with high credit scores are less
likely to have serious delinquencies than consumers with low scores. This section discusses properties of the score:
how it is built and how the results can be interpreted.
2.1 Segmentation
The predictability of a model is often greatly enhanced by segmenting the population into a number of subgroups,
and creating a different predictive formula in each segment.
Different formulas may be needed because there may be
differences in the availability of data for certain parts of the
population. For example, there is no need to have
attributes for public records in all formulas if there are
different segments for consumers with and without public
records. Another reason is that different business
decisions may apply, such as if companies have different
strategies for consumers who are new to credit. A third
reason for segmentation is that there may be certain
subgroups of the population for which there is a different
relationship between the modeling attributes and the
outcome.
Score Complete uses a segmentation scheme based on
delinquency and public records, the age of the oldest trade, and the number of trades on file. A total of eleven
segments are defined, of which eight are based entirely on credit file attributes, and three that are augmented with
Neighbourhood View data. A different formula is to be applied to each, so that the attributes can predict the outcome
over each segment. These distinct formulas are called scorecards. Since there is a direct relationship between the
segment and scorecard, the two terms are commonly used interchangeably.
Some credit files have robust data, with a long credit history and a large number of trades, and the future
performance of the consumer can be accurately assessed with great confidence. On the other hand, when the open
date of the oldest trade is recent and/or the number of trades is few, the consumer doesn’t have a robust credit
Consumers with good but short payment
histories may be considered low risk for
continuing payment and obligations with the
credit that they already have, but may not be
as low risk for new credit granted.
Identifying the segment for these consumers
will help a risk manager deal with these two
different cases.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 8
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
history and there isn’t a lot of information that can be used in identifying if these consumers are good credit risks.
These files are often called thin files. Consumers with good but short payment histories may be considered low risk
for continuing payment and obligations with the credit that they already have, but may not be as low risk for new
credit granted. Identifying the segment for these consumers will help a risk manager deal with these two different
cases.
In cases where aggregated data is used to augment the credit file, or is the only source of data when no file is found,
it may be good practice to be cautious in granting credit. While these consumers live in areas which suggest that
they may be good credit risks (and this has been demonstrated statistically), it may be wise to limit the exposure of
credit to consumers who have not personally demonstrated the responsibility of managing their own credit.
Applicants from the augmented scorecards can easily be identified and treated appropriately. This may depend on
the financial product. The applicant may be approved for a cell phone based on his or her address, or a credit card
with a moderate credit limit, but perhaps not for a large auto loan or a mortgage without a co-signer.
2.2 Modeling Technique
Eight of the eleven segments were built with individual credit file attributes (the RMS segments). The response for
each of these segments was a 90-day delinquency or worse (including derogatory public record) within the 12-month
performance window.
Within each segment, a logistic regression model was developed. Logistic regression is a modeling technique
designed to model the relationship between a binary
4
outcome and the explanatory variables. For each variable, a
weight is assigned, multiplying the weight by the value, or by giving a set number of points for each possible value of
the variable.
In addition to the logistic regression models, there are two neural networks. Neural networks are statistical models
that look at combinations of variables, and these combination variables are used to build the model. Neural networks
can be highly predictive when the event that is being predicted is a rare event or when the amount of available
information is limited.
With Score Complete, five of the eight segments enhance the predictability of the logistic regression models by
combining the logistic regression result with the result of the neural networks. The two results are blended using a
methodology called score fusion, transforming them into one estimate of the likelihood of serious delinquency. The
results were fused in such a way as to maintain the logistic regression as the primary driver of the score.
After score fusion, the formulas applied to each credit file have derived a probability of delinquency. These formulas
have been proven accurate, in the development dataset, by taking all of the records with similar probabilities and
calculating the observed bad rates and comparing them to the expected bad rate.
Three additional scorecards were used to create the score in cases where there was limited credit file information.
Recall from Section 1.1, all Canadian consumers with credit activity within the last 24 months and at least one valid
trade line will score based on their own credit file information. Consumers who do not qualify are most likely new
applicants for credit, and the response used in the development of these scorecards was a 90-day delinquency or
4
Binary outcomes are those that can take two values. These are often denoted by “yes” and “no,” or “true” and “false”, or in this
case, “good” or bad.” They are represented in code as 0 and 1.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 9
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
worse (including derogatory public record) within the 12-month performance window for any new account opened
within the first quarter of the performance window; trade lines which already existed on the file were not used in the
performance definition. In other words, the additional scorecards were built purely as an origination model.
Separate logistic regression scorecards were built for the consumers with limited credit file information and
consumers without any credit file at all
5
. Consumers who have a credit file, but the credit file has been created within
30 days and there is no trade on file, are somewhat similar to consumers who do not have a credit file at all. In order
to prevent a wild swing in score based on migration from the “no file” scorecard to the “thin file” scorecard,
interpolation between the two scorecards was done, based on the age of the file. This “hybrid” calculation is
presented as the eleventh scorecard.
Example: A consumer has a credit file that was created on March 1. At the time of an inquiry on March 1, without a
credit file, this consumer (applicant) would have to be scored based on his address and the aggregated data from
Neighbourhood View from his subdivision. He would be scored on the “No credit file found” scorecard, with a score
of 640. The inquiry opens a thin file. With his next inquiry, since he has a credit file, he can be scored on the “Thin
file augmented with aggregated data” scorecard. Using his address and thin file (with an inquiry on it), the calculation
of his Score Complete would be 670 on the “Thin file augmented with aggregated credit data” scorecard. This
consumer does not score 670 immediately after his file is created. His risk profile hasn’t changed very much, so his
score should not jump 30 points immediately. Instead, his score will increase by interpolation; 30 points in 30 days
means a point per day. His score on March 18, for example, would be calculated as:
Score Complete = No-Hit Score + (No-Score Score – No-Hit Score) x (Current Date – File Creation Date) / 30
= 640 + (670 – 640) x (18 – 1) / 30
= 640 + 30 x 17 / 30
= 657
This interpolation only applies for consumers with no trades on file, and with files created within 30 days. After 30
days, this consumer will be scored solely on the “Thin file augmented with aggregated data” scorecard.
2.3 Scaling
Score Complete uses a 300 to 900 scale rather than a probability estimate, where a high score indicates a low
delinquency risk. This is a convention that is standard to credit scoring. Most credit scores, regardless of bureau or
country or score version, follow the same scale. It is not necessary that the scores do this, but professionals who
work with different institutions that use different credit bureaus or scores, have gained experience in interpreting a
score value, and redeveloped versions and other scores tend to follow the same convention.
The key is that scores should separate and rank order delinquency risk. This means that there should be a
relationship between the score and the delinquency rate. Consumers with low scores have a high delinquency rate,
while consumers with higher scores should have a lower delinquency rate. The relationship should be consistent,
5
It may not be obvious how there can be records in the data where there is no credit file. The development data was selected by
finding all trade lines opened between July 2010 and March 2011. The performance of these accounts was determined at the
end of each of the three performance windows (June, September, and December 2011), giving each trade between 9 and 11
months of history. The observation dates of June, September, and December 2010 were used to pull information at origination.
If a consumer opened a new account in July 2010, they may not have had a file in June 2010. This consumer’s score would be
based on the address, and Neighbourhood View Street Level data from June 2010.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 10
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
and there should be a large difference in delinquency rates between consumers with the best and worst Score
Complete.
Score Range
Interpretation
300-499
Very serious issues, difficult to get any credit
500-574
High risk customer, may be required to provide securitization. May be eligible in the Telco industry.
575-649
Above average risk profile. May be granted credit with high interest or low limits.
650-749
Fairly safe credit risk for most institutions
750-900
Safe credit risk, generally approved easily
Scores below 300 or above 900 are not possible with Score Complete at Equifax Canada.
This interpretation should be treated as a guideline only. Companies should validate scores on their own portfolio
with their own definition of negative performance, in order that their risk managers can customize and optimize their
strategy for their own business needs.
Over time, changes in lending policies, reporting policies, lending institutions, consumer behaviour, and the economy
can change the relationship between scores and the expected bad rate. Score Complete will continue to perform in
the future and continue to rank order delinquencies, but the bad rate at various scores may shift if the data or the
economy changes significantly. Risk managers should monitor their portfolios regularly and determine if decisions
should be implemented at a different score range or cut-off than has been used previously.
The eight scorecards built with credit file information only, using the definition of a 90-day delinquency on any trade in
wallet were scaled identically; the same probability of delinquency in each scorecard was mapped to the same Score
Complete result on the 300 to 900 scale.
The additional scorecards were built with a different performance definition: 90-day delinquency on new trades. To
map to the same 300 to 900 scale and indicate the same degree of risk, two steps were taken.
First, new accounts opened between July 2010 and March 2011 for consumers for the development of the original
eight scorecards were analysed, and the relationship between the Score Complete result of these consumers to the
delinquency rate of these new accounts was determined. This was done to determine the relationship between the
Score Complete calculated from the original eight scorecards with the performance definition applied to the
augmented ones.
Second, an adjustment was made to the delinquency rate observed on the new open accounts used in the
development of the augmented scorecards. The reason this was necessary is because the development data
included all accounts for consumers who opened a new account, but is meant to apply to anyone who applies to
open a new account. Consumers who were approved for accounts are likely to have qualified after some additional
investigation by the lender, under criteria not accounted for in the data. It is likely that they are lower risks than
average, and the delinquency rates observed was lower than would be expected on a typical application. The
population used to build the model was not exactly representative of the population for which the model should apply,
and this correction was made to account for that difference.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 11
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
This adjusted delinquency rate was mapped to the delinquency of new accounts from the eight original scorecards
and the associated Score Complete. In this way, the final Score Complete result, 300 to 900, indicates the same
amount of risk regardless of which of the scorecards was used in the calculation.
Consumers with death notices on the file get a 300 score, the lowest possible score. The 300 score is reserved for
files with death notices. Instead of returning no score, which implies that the lender should undergo an investigation
of other sources to determine whether or not to approve credit, a death notice on file with a 300 score indicates that
no further investigation is needed. The only way to get a 300 score is with a death notice. All other cases will be 301
or higher.
2.4 Attributes
The credit file attributes that are included in the final Score Complete model are those credit file characteristics that
are found to be predictive of future delinquencies. These tend to be the same characteristics that risk managers
consider when they look at a credit file. They can be classified into a number of categories:
Payment history. If you are trying to determine whether a consumer will be able to make regular payments in the
future, the strongest predictor is to see if they have missed payments in the past. The number of accounts currently
past due, the rating of the most seriously delinquent trade, the number of accounts past due previously, and the
length of time since prior delinquency are considered here.
Utilization and balances. It is intuitively obvious that is it easier for consumers to pay their bills when they do not
owe a lot of money and the required payments are small. In addition, carrying balances on revolving accounts
suggests the inability to pay them off in full, which makes it more likely that the consumer will have problems
continuing to pay their bills. Low utilization indicates a difference between the credit limit and the balance,
sometimes called “open to buy.” A large amount open to buy gives consumers flexibility, a way to pay bills for a short
time that can’t be covered by income, by tapping into their lines of credit. A small amount of open to buy reduces the
incentive to make a payment on a credit card or line of credit since they won’t be able to use much of it anyway. And
once the credit cards are maxed out and the lines of credit are used up, they may not be able to cover their
payments. Bankruptcy may become their only option.
Credit history. A consumer who has managed credit for many years is considered lower risk than someone new to
credit. Over a long period of time, many people have significant life events, including moving, changing or losing job,
marriage or divorce, having children, and serious illness or injury. Those who have experienced these events in the
past and continued to pay their bills have a strong likelihood to continue doing so. They have shown responsibility
and consistency over a long period of time and can be considered a good credit risk.
On the other hand, young Canadians or new residents, new to credit, who have managed a first credit card with a
small credit limit have not demonstrated a history of managing large amounts of debt and haven’t proven the ability to
make regular payments over several years (such as a mortgage or auto loan). They are less likely to have dealt with
significant life events, and if one occurs, they have not demonstrated the ability to deal with it and maintain their
financial responsibilities.
Also in this category is the number and type of accounts. Consumers with many different accounts may be of higher
risk, and the properties of some trades indicate a higher risk than they do in other industries. For example, high
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 12
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
utilization in a line of credit may be a risk factor, while high utilization of a loan just means that it is a new account and
the consumer hasn’t had a chance to pay off much of the balance.
The presence of many new accounts may also be an indication of higher risk. For one thing, consumers in financial
difficulty often try to open new accounts in order to extend their “open to buy” and continue to pay their bills through
the tough times. Additionally, a number of new accounts may indicate that something has changed, and this brings
uncertainly to the risk prediction, which in turn means higher risk to the lender.
Public records. Those who have a prior history of bankruptcy, or have had collection issues or other derogatory
public records may be considered risky. The presence of these events, though relatively rare, has a significant
negative impact on a credit score.
Inquiries. Consumers who are going through financial difficulties, whether through job loss, family or health
situations, or general financial woes, often look for additional credit products to provide additional open to buy. They
may apply for a loan to pay down the credit card they have maxed out, and try to get a new credit card. The inquiry
may be the leading indicator, the first sign of danger that appears on the credit file. Of course not every inquiry is a
sign of financial difficulty, and only a number of inquiries, in combination with other warning signals should lead to a
significant decline in a credit score.
Consumers sometimes shop around when they are looking for certain products, and multiple inquiries over a short
period of time can be considered as shopping for one product. Three mortgage inquiries in a week rarely means that
a consumer is trying to buy three houses, while three credit card inquiries may mean that they are going to have
three new credit cards. Mortgage inquiries, auto finance inquiries, and Telco inquiries are deduped, meaning that
multiple inquiries within 30 days count as one inquiry in the calculation of the score. In addition, inquiries within the
first 30 days do not count in the score calculation, allowing consumers a chance to shop at different places without
there being an advantage to the first lender that pulls a file that receives a score with no inquiries counting.
The Neighbourhood View data elements that are used in Score Complete mimic the attributes for the individual.
Attributes such as the average number of delinquent accounts, the number and recency of public records, inquiries,
and high utilization are used, and have the same directional impact as they do at the individual level, although they
are weighted differently, and appropriately, for estimating risk.
In the case where a score can be generated from the credit file information only, Telco trades are not used in the
formulas. These scorecards use products including credit cards, loans, lines of credit, and mortgages from industries
which include banks, credit unions, and finance companies. The additional scorecards use the Telco trades, any
other trades in the file, inquiries, public records, and the aggregated data from the neighbourhood.
2.5 Account Management and Scorecard Migration
Score Complete is always calculated with the most up-to-date data available. Companies that want to monitor the
behaviour of their consumers often refresh the scores of their portfolio on a regular basis, and store the historical
scores in their databases. They may want to take action when consumers score differently.
When consumers are scored at different times, the information in the credit file will have changes, as new inquiries
and trades are added to the files and existing trades are updated. Even if there hasn’t been an update in the
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 13
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
information, the data in the credit file ages. Increasing the length of time since the oldest trade has been opened is a
positive factor, and negative behavior, such as missed payments and public records, move further into the past.
When Score Complete calculates a new score, it is based entirely on the current snapshot of the credit file (or the
archive snapshot for scores on a historical file). The consumer may score on the same scorecard, or may have
migrated to a different one.
When a consumer is scored on the same scorecard at two different points in time, the difference in the scores can be
attributed to changes in the attributes used by that scorecard. If the consumer’s attributes are similar from month to
month (no new trades, few inquiries, no large change in balances), the score will change very little. When attributes
change in a positive way (by paying down balances and keeping current on accounts and having delinquencies move
further into the past), the score will increase in a predictable manner.
When a consumer’s score is refreshed, and the consumer changes from one scorecard to another, the attributes
used in calculating the score change as well. Using different attributes with different weights sometimes causes a
change in the score that cannot be attributed to a single change in behavior. Moving from a thin file to thick (or to
very thick), from young to old, or from the augmented scorecard to the regular one may cause score changes that
can’t be explained by looking at individual attributes. Of course, moving from the Clean to Delinquent scorecards will
usually result in a much lower score.
Along with the score, companies may want to store the scorecard indicator in their databases. The scorecard
indicator is returned as Score Complete’s fourth reason code. When comparing the latest refresh to the saved
historical data, they would be able to determine consumers who have significant changes in score. They can further
understand these differences by comparing scorecard indicators. When the score change is based on scores from
the same scorecard, it can be said that there has been improvement in the underlying attributes. The consumer has
made positive changes in the characteristics of the credit file, and the score directly reflects that.
However, if the scorecard has changed, it is more complicated. There are two considerations. Sometimes a
consumer will migrate from a clean scorecard (92 to 95) to a delinquent one (96 to 99) – see table below. This is a
fundamental change in the file, and will usually be accompanied by a score decrease. However, changes between
thin and thick, or moving to the mature scorecards mean that different elements of the credit files are used in
evaluating risk, and so an increase or decrease in score is not necessarily attributed to improving or declining
attributes; it may simply be that the consumer has better or worse characteristics on the attributes considered more
important to the current segment than on the other scorecard. And movement from the augmented scorecards (89 to
91) to the established ones (92 to 99) indicates that the weight given to the neighbourhood transfers to the credit file
itself. It is expected in these cases that the score may change as consumers attain more robust credit files, and are
scored on different criteria. However, the scores should be correlated. For a population of consumers with scores
based on thin files and neighbourhood data, high scores imply that the risk is low. Low risk implies that consumers
are likely to make their payments on their new accounts. This positive behavior will lead them to have good credit
scores when they migrate to the traditional scorecards, and these high scores also predict the same expectation of
low risk.
Scorecard
Description
89
No credit file found
90
No trade on file and file age < 30 days
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 14
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Scorecard
Description
91
Thin file augmented with aggregated credit data
92
Clean, mature and very thick
93
Clean, mature and thick
94
Clean and thick
95
Clean and thin
96
Prior delinquency thick
97
Prior delinquency thin
98
Delinquent thick
99
Delinquent thin
3. Output
While the score is the most important output from the scoring module, several other pieces of information are
generated and returned with the output.
The first is a product identifier, “SC”, which identifies the score as being a Score Complete. Following the three-digit
score between 300 and 900 are four reason codes that help explain why a consumer file has the score that it does.
Reason codes correspond to the attributes that have values that lower the score, such as delinquent accounts, high
balances or utilization, or excessive inquiries. The first three reason codes correspond to the three attributes from
the credit file whose values impact the score by the largest amount. The fourth reason code denotes the scorecard,
or subpopulation, indicator.
The identification of the subpopulation used in the score may be a strong
indication of the score. For example, the subpopulations defined by
delinquency or public records tend to score lower than the subpopulations
without them. In addition, the scorecard indicator highlights files with limited
credit information, and the scorecards which are augmented with information
from the credit files in the neighbourhood. In many cases, these “thin files”
may be low risk for the accounts that they already have, but would be high
risk if they were given a new and large loan, line of credit, or mortgage. Risk
managers can use the scorecard indicator to automate different credit
policies for consumers belonging to different subpopulations.
There are only two cases where Score Complete will not return a valid score.
If there is no credit file, and there is no aggregated credit data from the address, there is no information available to
the score. The other is if a file has been flagged by Equifax as a manual file, or file under review
6
. Reject codes are
returned to account for these two cases.
See Appendix for a complete list of reason and reject codes for Score Complete.
6
This is a very rare occurrence. In one recent month, less than one inquiry in 500,000 resulted in a manual file.
Score Complete: 657
57 Too many auto inquiries
83 Too many derogatory public
records in neighbourhood
35 High credit open telco trades is
too high
91 Thin file augmented with
aggregated credit data
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 15
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
4. Using Score Complete – Alone or With Another Score
Score Complete is designed to score any credit file or application. It assesses the delinquency risk, either at
origination or in batch for account management purposes, and returns a three-digit score which rank-orders
delinquency risk. Score Complete returns a score virtually every time, unless there is problem with the input address,
such as a format error (address elements are in the wrong fields, or the address is out of country) or an address in a
new postal code that has just been created. Score complete rank-orders delinquency risk for all files, and can be
used by itself as a complete delinquency score.
Some companies already use a delinquency score, such as BEACON, Equifax Risk Score (ERS), or Consumer Risk
Predictor (CRP). These scores assess delinquency risk, but have minimum scoring criteria that do not score certain
credit files, and they certainly do not return a score when no credit file is found. These companies may want to
continue to use their choice of score, but have a waterfall process that uses Score Complete in cases where their
primary score fails to return a score. This can be done.
To do this, the primary score must be activated for the inquiry, as well as Score Complete as a child score. If the
primary score succeeds in scoring the file, the score identifier, three-digit score, and four reason codes are returned,
as well as a blank reject code. The Score Complete algorithm does not run, and no result is returned.
If the primary score fails, the score identifier is still returned, along with a 0 score that indicates no score can be
calculated. The reason codes will be blank. The reject code will be filled in, corresponding to the code for that
particular score (reject codes do vary between the different generic scoring models).
Then the Score Complete algorithm runs, and generates a score. The SC score identifier, the three-digit score, and
four reason codes are returned. The fourth reason code is the scorecard indicator. In the case of no data, neither a
credit file nor aggregate data is available, the reject code will be returned.
Score Complete’s 300 to 900 score range is similar to the range used by most versions of most of the generic
delinquency scores available from Equifax. It may be appropriate in some cases to use the same strategies with a
Score Complete result in the second bureau segment as would be used with a score in the primary segment,
although companies would be wise to test this assumption on their own before relying on that assumption.
5. Evaluating a Model
What do we mean when we say that a model works? It means that a model can be used in predicting the unknown
quantity that it is supposed to predict. There are a number of different ways to determine this. Formal methodologies
include creating tables and graphs and calculating statistics.
Credit scores, especially with scaling, work if the scores separate and rank order credit risk. A population that is
representative of the target population that has values for the required quantity are used to show how well the score
predicts the outcome.
First, records are segmented by score. This may be done by fixed score ranges, like 20-point score bands, or by
score distribution, such as deciles. The keys to looking at data expressed this way:
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 16
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Separation. Files with low scores should have a high percentage of records with the negative occurrence
that was measured, while low risk scores should have a low percentage of bad outcomes. The more
separation there is, the better the score is performing. It is better if the bad rate is 50% in the high risk
section and 2% in the low risk, than if the differences vary between 15% and 10%.
Rank ordering. As scores improve, the bad rate should improve in an orderly and predictable fashion.
Score Complete scores increase when the risk decreases, so the observed bad rate should also decrease
as scores increase.
Here is an example illustrating how scores separate and rank order risk:
In this example, the records are grouped in deciles, 10% in each row of the table. Within the entire population, only
3.6% of the records were classified with negative performance (792K out of 22M), having a 90+ day delinquency or
bankruptcy within the performance window.
The lowest scores, 300 to 650, correspond to the 2.2M consumers (10% of the total) with the highest risk. They had
a 21.8% bad rate, which is 6 times higher than the overall bad rate. The highest scores, 844 to 900, had a very low
bad rate of 0.4%. The bad rates rank order through the score ranges, decreasing with every decile.
Bad rates are often shown graphically. This is a plot of the column represented in the table above by the red up
arrow:
All Accounts
Negative Performance
Delinquency 90+ or bankruptcy
21.8%
5.9%
2.8%
1.6%
1.1%
0.8%
0.6%
0.5%
0.5%
0.4%
0.0%
5.0%
10.0%
15.0%
20.0%
25.0%
300 to 650 650 to 701 701 to 730 730 to 752 752 to 772 772 to 790 790 to 809 809 to 825 825 to 844 844 to 900
Separate and rank
order
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 17
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Companies will often set a score cut-off, with strategies to take adverse action to consumers with low scores, or will
decline applicants with low scores that fall below the cut-off that has been set. Since they want to limit declines or
adverse action to a small percentage of their consumers or applicants, it is crucial that most of the consumers who
will, in the future, exhibit bad behaviour have scores that are below the cut-off now. Some good customers
7
have low
scores for various reasons, and companies do not want to strain many of these relationships and lose good business.
In the table above, of the 792,361 consumers with serious delinquency or bankruptcy, 482,191 scored in the bottom
10%. This represented 60.9% of all bads. It is often a key measure of the predictability of a credit score to determine
the percentage of bad records identified in the highest risk scores, and the higher the better. This is the column in
the table above highlighted by the purple down arrow.
Graphically, this is called a lift chart:
The better a score works, the higher the line will be in the lift chart. One way to measure how well the score works is
called the Kolmogorov-Smirnov statistic, or K-S. The K-S looks at the lift chart, and calculates the percentage of
good and bad records (represented by the blue line, below) scoring below each available score value. The maximum
difference is the K-S
8
:
7
Good means that they will be good customers in the future; they will meet their financial obligations, and not be classified as
bad according to the definition that is used to test the score.
8
Other metrics, such as the Gini and AUROC, measure the lift of a model differently, but K-S is often the standard metric in
credit risk scoring.
0%
61%
77%
85%
90%
93%
95%
96%
98%
99%
100%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
300 650 701 730 752 772 790 809 825 844 900
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 18
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Here, 77% of the bads (red line) score at 701 or below, but only 18% of the goods. The 59% difference is the largest
difference at any point on the graph, so it is the K-S value. The higher the K-S, the better
9
.
5.1 Score Complete Development Data
Score Complete is built in two pieces, the traditional credit score that is based on predicting delinquency from records
with robust information, and a separate component for the no-hits and thin files. Traditional credit scores have been
developed and redeveloped for many years by different modellers, and these scores have been accepted as valuable
and predictive.
The concept of scoring every application, including no-hits and thin files may need some additional support. The
results below are based on the data from the validation dataset from the score development data
10
. A bias
correction, as described in section 2.3 to adjust the delinquency rate prior to the model scaling, is incorporated here
as well.
The scorecards based on no-hits and thin files, applied to consumers who opened a new account between July 2010
and March 2011, show significant separation in the prediction of 90-day delinquencies:
9
What is a good K-S? It depends greatly on the use of the model and the available data. Sometimes a very small amount of lift
can have tremendous benefit. One of the values of K-S is as a comparison tool between different models on the same data, as
an evaluation of the models. Another is to evaluate how well a model performs over time as things change.
10
Traditionally, when statisticians create a model, they separate the initial dataset into two parts, often with approximately the
same number of records in each. The training dataset is the one used to build the model, and create the weights for each
attribute. The formulas are optimized on the training dataset. The validation dataset is used to show that the formulas work on
different data. It is fair to use the validation dataset to make inferences on how well the model would work in independent
analyses.
0%
61%
77%
85%
90%
93%
95%
96%
98%
99%
100%
0%
8%
18%
28%
38%
48%
59%
69%
79%
90%
100%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
300 650 701 730 752 772 790 809 825 844 900
59%
K-S
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 19
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
The K-S is 37.2%. Given that there is limited, if any, credit file information, and the outcome is based on opening
new accounts rather than managing existing ones, this K-S is very good.
While the traditional scorecards were developed with a negative performance definition of a 90-day delinquency or
worse on any trade, not just new ones, the new accounts for consumers on these scorecards were analysed to
estimate the relationship between the score and the probability of delinquency on a new account. As the following
chart demonstrates, the bad rate by score range for the traditional scorecards and the augmented ones indicates the
same expectation of a 90+ day delinquency on a new account.
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
301
to
588
588
to
610
610
to
627
627
to
640
640
to
651
651
to
660
660
to
668
668
to
674
674
to
677
677
to
682
682
to
686
686
to
692
692
to
697
697
to
701
701
to
706
706
to
717
717
to
729
729
to
741
741
to
752
752
to
900
0%
10%
20%
30%
40%
50%
60%
70%
500 600 700 800 900
Bad Rate
Score Complete
Calibration of New Scorecards
(Accounts Opened First Quarter, 12 month performance)
Augmented Original
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 20
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
These two charts show that the Score Complete augmented scorecards act the same as the original ones. They
separate the high risk applications from the low risk applications and return a score that is consistent with the
expectation of delinquency. Risk managers may be comfortable using the result of the Score Complete model to
help assess the likelihood that a given applicant will make regular payments on their financial obligations, and will be
pleased that so many applications are assessed where traditional credit scores fail.
Summary
Canadian consumers need credit cards for everyday purchases, loans to buy goods and services, lines of credit to
optimize and bring flexibility to credit management, and mortgages so they will have a place to live. To finance these
things, banks and other lending or credit granting institutions provide these credit products, but need to ensure that
they can do so profitably. They send their data to Equifax, and Equifax builds credit files that contain the credit
history of over 24 million active credit holders.
The information in these credit files allows risk managers the opportunity to evaluate the likelihood that the consumer
or applicant is in a good financial position and will be able to repay the loan, or make regular payments on the
revolving credit. A credit score, like Score, assists this process. Score Complete weighs the elements of the credit
file properly, and provides consistency to the decision making process, allowing the risk manager the ability to
optimize strategies to be as profitable and successful as possible.
The fact that Score Complete succeeds in returning a score on virtually every request makes it an outstanding risk
management tool for any lender.
Additional Reading
The original eight scorecards used when a file is robust are the same as the Equifax Risk Score (ERS 2.0), with the
only exception being that ERS 2.0 scores of 300 are bumped to 301 to free up the 300 score to account for death
notices (Death notice files do not score with ERS 2.0). The reason is referred to the Equifax Risk Score 2.0
Handbook.
As the additional scorecards are somewhat based on Neighbourhood View, for a further understanding of that
product, please see the Neighbourhood View Handbook.
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 21
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Appendix – Reason Codes, Scorecard Indicator, Reject Codes
Reason Codes
Reason
Code
Description
1
Average trade age too young
2
Too many trades currently 90+ DPD
3
Too many trades 60+DPD within the last 2 years
4
Too many trades opened within last 2 Years were 90+DPD
5
Too few satisfactory trades
6
Too few open trades
7
Too many trades past due
8
Too few trades on file
9
Length of time trades established
10
Percentage of trades satisfactory too low
11
Current delinquency
12
Delinquency on file
13
Too few auto trades
14
Too few satisfactory auto trades
15
Too many auto inquiries
16
Too few bank instalment trades
17
Too few satisfactory bank instalment trades
18
High ratio of balance to high credit on bank instalment trades
19
Current delinquency on bank instalment trades
20
Delinquency on bank instalment trades
21
Too many bank inquiries
22
Too few bank revolving trades
23
Too few bank revolving trades older than 6 months
24
Length of time bank revolving trades established
25
Too few satisfactory bank revolving trades
26
High ratio of balance to high credit on bank revolving trades
27
Current delinquency on bank revolving trades
28
Delinquency on bank revolving trades
29
Too many collection inquiries
30
Recent collection inquiry
31
High balance on collection items
32
Too many collection items within the last 3 years
33
Too few satisfactory credit union trades
34
Too many finance instalment trades
35
High credit open telco trades is too high
36
High ratio of balance to high credit on finance instalment trades
37
Too many personal finance inquiries in the last year
38
Too many personal finance inquiries
39
Too many finance revolving trades
40
High ratio of balance to high credit on finance revolving trades
41
Delinquency on instalment trades
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 22
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Reason
Code
Description
42
Too many national credit card inquiries
43
Average national credit card trade too young
44
High balance on national credit cards
45
Too many national credit cards
46
Too recent age of oldest trade in neighbourhood
47
Too many national credit cards with high utilization
48
Too many national credit cards with high utilization
49
Too many national credit cards with high utilization
50
High ratio of balance to high credit on national credit cards
51
High ratio of balance to high credit on new national credit cards
52
Current delinquency on National Card Trades
53
Delinquency on National Card Trades
54
Too many other inquiries
55
Recent public record
56
Public records within the last year
57
Public records within the last 3 years
58
High ratio of balance to high credit on personal finance trades
59
Too many revolving trades with balances
60
Too few revolving trades with $0 balance
61
Length of time revolving trades established
62
Too many revolving trades with high utilization
63
Too many revolving trades with high utilization
64
High ratio of balance to high credit on revolving trades
65
High ratio of balance to high credit on new revolving trades
66
Too many retail trades with balances
67
High ratio of balance to high credit on retail trades
68
High ratio of balance to high credit on sales finance trades
69
Current delinquency on sales finance trades
70
Too many telco inquiries in the last year
71
Too many telco inquiries
72
Too many inquiries in the last year
73
Too many recent auto, sales finance, national cards or other inquiries
74
Too many inquiries in the last 3 months
75
Too many inquiries
76
Too many auto, sales finance, national cards or other inquiries
77
Recent inquiry
78
Too many revolving trades with high utilization
79
Too many department stores inquiries on file
80
Current maximum rate in neighbourhood too high
81
Too few open trades in neighbourhood
82
Too many delinquent trades in neighbourhood
83
Too many derogatory public records in neighbourhood
84
Too many inquiries in neighbourhood
85
High utilization on open revolving trades in neighbourhood
86
Too few satisfactory trades in neighbourhood
Confidential and Proprietary
©2014 Equifax Canada Co. All rights reserved. Page | 23
Equifax®, Score Complete™ and Neighbourhood View™ are registered trademarks of Equifax Canada Co.
Reason
Code
Description
87
Too low high credit on open trades in neighbourhood
88
Death notice on file
Reject Codes
Reject Code
Description
G
No credit file or neighbourhood data available
F
Score Complete not available, file under review
Score Card Indicators
Scorecard
Description
89
No credit file found
90
No trade on file and file age < 30 days
91
Thin file augmented with aggregated credit data
92
Clean, mature and very thick
93
Clean, mature and thick
94
Clean and thick
95
Clean and thin
96
Prior delinquency thick
97
Prior delinquency thin
98
Delinquent thick
99
Delinquent thin