Credit Score Guide
New for 2006
As of January 2006 Bogeyman Credit will be using the latest risk scoring developed as a joint venture between Professor Nick Wilson, Director of the Credit Management Research Centre (CMRC) and ICC Credit, one of the largest providers of Corporate data in Europe.
This is a range of highly predictive new score cards for UK commercial credit risk assessment which will further strengthens ICC Credit’s already formidable reputation for predictive modelling. All UK limited companies will be assessed for scoring – including companies which have not filed accounts.
A highly predictive new model will be used to score the extensive unincorporated database.
More predictive scoring
Scores were calculated for a large sample of failed and non-failed businesses one year prior to the failure event in order to determine how well it would ‘predict’ the event. Further diagnostic statistics such as the Kolmogorov-Smirnov statistic and the Gini coefficient were used to verify the high predictive accuracy of the score. The result is that the score is even more predictive than the existing score in analysing whether companies are going to fail within 12 months.
• The new score for unincorporated companies is 18% more predictive than the existing score
• The new score for limited companies is 15% more predictive than the existing score.
The predictive accuracy of the score is constantly monitored in relation to all incidences of business bankruptcy and insolvency.
Other features of the new ICC Risk Score
• All UK companies have been assessed for scoring.
• All UK companies have been assigned a credit limit.
• Extensive online database of credit limits for UK companies and non-limited businesses.
• New data used to score non-limited businesses including socio-economic, demographic and industry risk weightings.
• A consistent scoring range of 1 to 100 has been assigned to limited and nonlimited.
• The commentary on why and how a business has been scored has been expanded for limited companies and is now provided for all non-limited businesses.
• Scoring criteria have been enhanced to provide an accurate analysis of risk
What is a scoring model?
Commercial risk scoring involves developing scorecards that summarise risk information. Score development is a structured and scientific process with the first stage defining an objective for the model or an ‘outcome’ that has to be modeled.
In the case of commercial risk scoring, the objective is to understand whether a business might experience financial difficulty and ultimately fail (become insolvent or bankrupt).
ICC Credit’s scoring model
In the course of any one year, approximately 2% of trading companies will become insolvent. ICC Credit has sought to establish cards that offer the most robust predictive scores in the UK to help prevent the extension of credit to ‘bad’ companies.
For ease of use and consistency, the score for each company is presented in the range of 1-100, where 1 represents the highest risk and 100 the lowest risk. Risk is determined as the probability of the company becoming insolvent or bankrupt within the next 12 months. Certain events will trigger an immediate suspension of the score and associated credit limit. This includes notification of insolvency or bankruptcy proceedings and significant CCJ information.
How is the score banded?
| Limited companies |
| 1-35 |
Caution. |
High-risk potential. |
|
| 36-50 |
Caution. |
Moderate Risk Potential. |
Measured exposure. |
| 51-60 |
Normal. |
Limited Risk Potential. |
Normal terms. |
| 61-100 |
Confidence. |
Low Risk Potential. |
|
| Non-Limited companies |
| 1-43 |
Caution. |
High-risk potential. |
|
| 44-57 |
Caution. |
Moderate Risk Potential. |
Measured exposure. |
| 58-69 |
Normal. |
Limited Risk Potential. |
Normal terms. |
| 70-100 |
Confidence. |
Low Risk Potential. |
|
What is the Credit Limit?
The credit limit is an absolute measure of a company’s ability to settle potential credit transactions at any given moment.
Example of top-level criteria used to risk score limited companies
| Demographic |
Financial |
Trading |
| Age of Business |
Financial Ratios |
Court Judgements
-Number
-Timing
-Value |
| |
|
|
| Is there a parent company |
Financial Trends |
Insolvency Notification |
| Industry Risk Weight |
Relative Ratios |
|
| Legal Form |
Net Worth/Size |
|
| |
Cash Situation |
|
| |
Filing history |
|
| |
Audit Qualification
– type |
|
| |
|
|
Credit Limit for limited companies
The credit limits (which are now capped at £500 million with a minimum of £500) are calculated as a percentage of absolute balance sheet and cash flow value. Credit limits will be adjusted for County Court Judgement information. The resulting limit should be regarded as a yardstick for the possible level of acceptable credit, or alternatively the maximum amount to be owed by an applicant (customer).
Example of top-level criteria used to risk score newly incorporated companies:
| Demographic |
Financial |
Trading |
| Age of Business |
Filing History |
Court Judgements
-Number
-Timing
-Value |
| |
|
|
| Industry Risk Weight |
Insolvency Notification |
|
Credit Limit for newly incorporated companies
Depending on the legal status of a company an initial limit of between £500 and £5,000 is given. This initial limit then increases over the course of time but can be revised downwards if adverse data is filed. Once a set of accounts is filed, the normal methodology for limit calculation applies.
Credit Risk Score model for non-limited companies
The ICC Credit risk scores for non-limited businesses were developed by analyzing data from over 18,000 small businesses over a period of 4 years. The model was structured to predict the eventuality of the business becoming bankrupt within the next 12 months. The database was constructed from the ICC Credit database of non-limited businesses plus data feeds of industry level failure rates and postcode level data. The latter database included socio-economic and geodemographic information, indicators of local economic conditions and incidences of bankruptcy and CCJ’s. The database had information on over 1.8 million individual postcodes.
The scoring model was developed using logistic regression analysis and analysed over 200 potential predictor variables before specifying the final model. The model was validated on the live sample of ICC Credit non-limited businesses.
The lack of disclosed financial data is overcome by the use of ICC Credit company industry data which is used to estimate the financial strength of the non-limited business.
Example of top-level criteria used to risk score non-limited companies
| Demographic |
Financial |
Trading |
| Age of Business |
Number of employees |
Bankruptcy Notification |
| |
|
Court Judgements
-Number
-Timing
-Value |
| Industry Risk Weight |
|
|
| Premise Type |
|
|
Credit Limit for non-limited companies
The credit limit is calculated by taking a percentage of the estimated strength. The percentage to apply is weighted in accordance with the relative risk, as determined by the score. The higher the score (hence lower the risk) the greater the percentage to apply. Credit limits are now assigned to all geographical areas in the UK and Northern Ireland. Credit limits will be adjusted for County Court Judgement information.
|