GCD's Mission is to help banks understand and model credit risks. The comprehensive data pools are collected over a decade and distributed back to members for their own research and modelling.

 

Learn More

GCD is a unique data consortium that owns banks internal data for both PD and LGD. GCD’s data pools support the key parameters of banks’ credit risk modelling: Probability of Default (PD), Loss Given Default (LGD), Exposure at Default (EAD).

Learn More

GCD’s library gives access to wide variety of publications on risk related topics. Global Credit Data members work together to analyse the data and discuss methodology issues. GCD has published numerous papers and is actively promoting academic research on the data collected.

Access the Library 

Members not only benefit from exclusive rights and access to credit databases and analytics, but also from knowledge and research facilitation possible via the unique industry association.

Through a variety of forums such as workshops, webinars and surveys, GCD is an active industry participant facilitating the discussion in key strategic areas.

Learn More

Global Credit Data collects raw data from its members and distributes it back to them for use in their own analysis and modelling. GCD supports its members by providing a flexible high-end tool on the data pool: the GCD Visual Analyzer. Member banks can create dynamic Reference Data Sets and generate instant views on the data.

Learn More

-> menu code <-

Frequently Asked Questions

Global Credit Data Membership

GCD Service Offerings

GCD offers a wide variety of services, that are driven by data submission from member banks. 

Learn more on how GCD Data will benefit your bank, contact us at secretary@globalcreditdata.org.

 

Data Quality

Below is a list of frequently asked questions focused on the data quality aspect of GCD’s work:

Does GCD audit the data systems at each bank?

No, that would be too intrusive. GCD relies on the member-banks internal audit and validation processes. Global Credit Data can't go into the banks and audit their data entry from source systems or paper records but we can look at the data from all angles possible. GCD uses automated validations of the data fields and values in their input portal as well as in-cycle audit where a member’s data is manually audited before aggregation, looking for biases, bad data, etc.

Does GCD clean the data after receipt?

No, each bank does its own cleaning during the collection process to pass the validation and audit. GCD never changes values or removes data, but we do advise members to do this when we see a need for correction.

Could a member supply false data?

Yes, but this would require collusive fraud and any validator or auditor would be able to see the returned data and check that it was not real. The return of data to each member has a specific field marking which data has been submitted by that bank. One purpose of this is to allow the auditors, validators and regulators to check this against the known good records of that bank. In other words, GCD data is as good as the data banks are using internally.

Can GCD reject data which passes all validations?

Yes, GCD routinely questions incoming data for comparability and requests banks to reconsider. GCD also completes annual out-of-cycle audits of members’ complete data sets in order to ensure completeness and compatibility with other banks’ data. A scoring system including feedback is used to let members understand their own data quality in comparison to others.

Do banks change the data after a GCD audit?

Yes, by comparing data from one member with others, GCD can point out likely data collection errors and members usually fix these quickly. Peer Comparison reports are a new tool introduced to compare key statistics on the data of the submitting bank vs other banks. The above processes ensure that member banks are aware of any problems with the quality of their data and are able to solve them in order to make the most out of their membership with Global Credit Data.

What happens to old data after rule changes?

We try to automatically adjust fields for old data when a field or definition changes, but many times we request for members to re-submit all data (3-year cycle).

Data Submissions

GCD has a recognised track record re data pooling while ensuring data quality, security and confidentiality.

How do I prepare the data submission and which data is necessary to send ?
Our User Guide section has the documentation you are looking for.
How has GCD set up its data collection process and ensures the appropriate level of data security?

Global Credit Data engages an independent Data Agent to assist in collecting and sorting the large quantities of data and the ongoing monitoring of data contributions. The Data Agent is Capgemini working in collaboration with SAS, who supply the software platform and aggregation tools. Cap Gemini also ensures anonymity of contributions and undertakes standardized analyses and reports for the benefit of all member banks. Data will be transmitted only by encrypted and secured protocols and processed by mature data portal with embedded validation rules. Our data transmission hosting platform is tailored to the requirements of our financial services contributors.

How can member banks submit data?

Members can do a submission during the submission cycles ("in cycle submission") or - after discussion with GCD Executives - outside the submission cycles ("out-of-cycle submission"). In order to do a submission, the members are preparing their data first in the appropriate format and then upload the data on to the data portal https://www.globalcreditdata.net. Members can run as many trial submissions as they want. In case they choose a firm submission, the data will be checked and audited by GCD Executives (see for more details the FAQs on data quality and our Data Quality Approach).

How can I get access right to the dataportal?

Please contact Hale Tatar or Michael Dhaenens.

My browser does not show any information when I click on the link to the dataportal?

In order to ensure that only member banks can access the portal, GCD and Capgemini make use of the IP whitelist method. This method involves creating a list of the public IP addresses used by each member bank and then checking to make sure that for each logon attempt at the portal it is coming from inside a member bank. Please check your IP adress (for more information see our IP whitelist information) and communicate your IP adress to our Cap Gemini support: globalcreditdata-support.bnl@capgemini.com

 

cap_gemini_new_logo.png

Link to the data portal:

https://www.globalcreditdata.net

LGD & EAD Platform

For what purposes can I use the data from the LGD & EAD platform?

Participating member banks use our database both for modelling and benchmarking purposes. Although banks are required to use their own internal default history to build and calibrate LGD models, in case of low default portfolios additional data information can be necessary. As GCD is collecting a broad range of information for each defaulted borrowers, banks have the possibility to create their own reference data set, representative for their banks' portfolio. GCD also publishes peer comparison reports which give banks further insights into their models.

Some concrete use cases are:

  • Identify risk-drivers on a more diverse dataset (e.g. Segmentation, LtV, Time-to-Recovery, …)
  • Identify macro-economic dependencies of LGD and EaD
  • Prove the correct LGD levels for Low Default Portfolios (e.g. banks, shipping)
  • Correctly calibrate downturn or stressed LGDs from long time series
  • Reduce uncertainty add-ons for lack of data
  • Benchmark historical losses with your modelled forward-looking expected losses under IFRS 9 / CECL
  • Peer benchmark the LGD estimates underlying your pricing models with loss rates from a global and diverse dataset

The database may contain data on the same defaults collected from different member banks. How does GCD account for possible repetitions? How does GCD recommend their member banks to treat the data to eliminate the threat of possible repetitions?

Although Banks may have exposures to the same Borrower these should not be viewed as duplicates as each Bank will have different types of exposure with that Borrower and will give each Lender a different default experience.

Does GCD only collect resolved defaults or also unresolved defaults?

GCD does collect both resolved and unresolved defaults. Banks who submitted both resolved and unresolved defaults, will receive back both resolved and unresolved defaults. Banks who only submit resolved loans, will only receive bank resolved loans.

How do you define a case as resolved vs unresolved? Could you share the definition of the same?

In general terms, we want to classify only those cases as resolved where the bank delivering data is satisfied that the final financial outcome of the default is known and it is therefore possible to calculate the final LGD without much likelihood of it being changed by subsequent data.  A Borrower is only to be regarded as resolved if all of the loans associated with it are resolved.

PD & Rating Platform

For what purposes can I use the data from the PD & Rating platform?

By making use of the database, participating member banks have the possibility to further enhance the backtesting, calibration and benchmarking of their internal PD models. Although banks are required to use their own internal default history to build and calibrate PD models, in case of low default portfolios additional data information can be necessary to ensure a smooth calibration or backtesting. In addition, members banks can directly compare their level of calibration and model accuracy with peers and therefore get further insights into the overall performance of their models.

Some concrete use cases are:

  • Benchmark your PD masterscale (comparison PDs and default rates per rating class)
  • Benchmark your system’s discriminatory power
  • Identify macro-economic dependencies in default and migration data: Extract a “systemic factor” from rating migrations or default rates
  • Benchmark your asset correlations and long term default rates
  • Benchmark your stage allocation / SICR buckets (thresholds for “life-time PD” movements) under IFRS 9
  • Reduce uncertainty add-ons for lack of data

How does GCD take into account that the participating member banks use different methods on how to calculate default rates?

Many market practitioners, such as portfolio owners (banks and investment companies), external rating agencies, auditors/ regulators and academics perform and publish calculations on default and transition rates. For estimating transition and default probabilities of rating grades, the so-called “cohort approach” is still considered to be the most common one. The committees of GCD have decided to implement that approach for its “PD & Rating platform” in order to get uniform results and results which are the most comparable with other calculations from market participants.

Members banks deliver entire portfolio default information at (anonymized) borrower level and allow GCD to calculate default rates and migration matrices on the defined GCD methodology.

The defined methodology takes into account how to deal with “double defaults / multiple defaults”, exits from the portfolio, new borrowers etc. and the aggregated output is, therefore, uniform for all participating banks. Nevertheless, further analysis is possible for all participating banks on the effect of those elements, using the additional markers delivered.

How does GCD take into account that the participating member banks use different internal rating scales?

The requirement is that banks map their internal rating system categories to the S&P rating scale. Therefore, the delivery to GCD does not include the raw internal ratings but the “internal rating mapped to the nearest external rating category”. Banks need to do this mapping for other purposes already, such as external disclosure, regulatory reports and even for using rating agency historical default data and therefore have the knowledge in house to perform such a mapping before submitting data to GCD. Our advice is that they refer to the rating agency’s guidance on definitions of rating grades as well as take into account the published long-term default rates of the rating agencies.

Benchmarking Platform

*NO INFO* FAQs on the Benchmarking platform

More information coming soon ....

Working Groups

How do working groups get started?

A: GCD has a motto “by banks for banks” and nowhere is this more true than in the case of working groups. GCD’s Methodology Committee has a rule that working groups can only start when a member of the Methcom (or the Board) is prepared to chair it.  This ensures that the money spent on consultants and executives is used for research or support in areas which members want. Any member can apply to join the Methodology Committee and any working group at any time. If you have an idea which you think could benefit from a working group approach then please raise this either with executives or the chair of the Methcom.

How do the working groups work?

In the startup phase of a working group the Chairman and members decide on their goals and deliverables. If the deliverables involve serious analytics or IT work then the executive will organize to find bidders for this through an RFP process.  If the working group mainly needs to change the data template or documentation then this will be agreed during meetings and draft documents prepared by executives. Most working groups have between 10 and 20 persons registered as members, but due to time constraints not everyone attends every meeting. Regular presentations are made at the Methcom and at each General Meeting to keep all members updated.

What do working group members need to do?

Usually meetings are held once per month although sometimes more often during critical times or less often when members are strongly involved in other activities.  The meetings are done using WEBEX, with an agenda and invitation and usually some preparatory material sent to each member. The meetings are usually recorded on WEBEX and notes of decisions are always taken so that if a member misses a meeting then they can catch up. Members are not required to have any special skills in the area under discussion and it is understood that not everyone has time to pre-read everything. Some banks have multiple members in a working group and some banks only ever send one person.  It is up to you.

hy do we have focus groups (next to working groups)?

Focus Groups have been created by Methcom to allow Banks to discuss topics of interest at an informal level. It is necessary for Focus Groups to have an agreed upon topic but they are not required to define objectives/outcomes and no deliverables are expected.·Focus Groups do not have the ability to bring changes to the template to Methcom but can bring suggestions. Global Credit Data Executive time will be limited to 10 hours a month and can involve setting up webex meetings, taking short notes and performing quick analytics. Focus Groups must be open to participation by all Member Banks.

How can I establish a focus group?

Any member bank can suggest a Focus Group to Methcom or executives. If Methcom agrees then the Focus Group is established and run by Member Bank with limited assistance by Global Credit Data Executives. If the Focus Group wishes to proceed further then they will be required to request for a budget and request for more Executive Time at Methcom.  If agreed by the board (subject to its mandate from the GM) then the Focus Group would change to a Working Group with pre-defined and agreed upon objectives and deliverables.