Submission - Credit Reporting Regulatory Framework: Submission to ALRC Privacy Inquiry (December 2007)

[« previous] [next »] [up] [title page] [full contents]

2.3. Quality and accuracy of data

Concerns about the quality and accuracy of data include name mismatches, re-listed data, multiple listings, and disputed data. Quality concerns might also include the age of the data, as there are questions over whether old data is indicative of current risk to lenders.

It is the view of this Report that accuracy of data in the credit reporting environment is one of the most important consumer issues when analysing potential consumer harm. The consequences of inaccurate credit reporting information are significant.

Accuracy concerns have also been an obstacle to the industry’s efforts to have privacy law reformed in order to allow more comprehensive reporting.

Credit reporting agencies have made some limited internal assessment of the accuracy of credit reporting information by assessing a selection of sample files. This testing indicated that around 1% of sample credit reporting information files contained significant errors. Around 4% of sample files contained minor errors that are unlikely to have negative consequences for consumers. Around 95% of files were error free.[8]

Industry representatives note that this level of accuracy is acceptable when the volume of data and transactions is taken into account:

Despite the anecdotal evidence to the contrary, independent research demonstrates that the data quality is very high given the highly transactional nature of the data base with over 80,000 real time transactions a day.[9]

Consumer and privacy advocates have expressed significant concern over the accuracy of credit reporting information and its potential consequences. Although no comprehensive information on data accuracy has been made available, consumer and privacy representatives have identified accuracy problems through individual complaints,[10] systemic complaints,[11] previous OPC audits,[12] and small consumer surveys.[13]

Fortunately, a wide range of solutions to data accuracy issues in credit reporting is available. Many of these are discussed in detail in DP72. The following summary is not exhaustive:

• Greater consumer involvement
  It is anticipated that any increase in consumer involvement in monitoring credit reporting information will lead to an improvement in data quality as consumers recognise mistakes in their own files. This was the experience in the small CHOICE survey where consumers quickly recognised basic errors once they obtained copies of their own credit reports. Improving consumer involvement in credit reporting is discussed in greater detail in this Report at Section 2.4 (page 12).
• More timely notice of listings
  Consumers are in the best position to recognise the accuracy of data being added to their credit report. However, they are currently not always consulted at the key times when information is being added to their file – such as the listing of a default. It is essential to have listings sent to consumers to check while the information is fresh. It is difficult for consumers to recognise and correct inaccurate information many years later. Improvements to the notice requirements for credit reporting information are discussed in more detail in this Report at Section 2.2 (page 7).
• Correction of data
  It is important for disputes and ‘notes on file’ to have a real impact on the accuracy of credit reporting information. In practice it appears that disputed information and notes on file do not receive due consideration in the credit scoring process and they may not be seen by potential credit providers. An effective mechanism for correcting inaccurate data and noting disputed data must be delivered.
• Removal of old data
  Despite the intention of the existing Part IIIA rules to exclude statute-barred debt and old defaults (more than five years old) from appearing in credit reporting information, there have been some circumstances where this information has remained on the credit report with negative consequences for consumers. This issue is discussed in detail in DP72 and is likely to be the subject of tighter rules on the listing and re-listing of defaults.
• Removal of duplicate data and multiple listings
  There are significant concerns that credit reports currently contain multiple listings and duplicate data relating to the same defaults.[14] Some submissions to the ALRC have suggested that a mechanism should be included for updating a listing rather than having multiple listings appearing on an individual’s report.
• Audits
  Regular audits of the accuracy of data could play a significant role in improving the accuracy of credit reporting information. Industry-initiated self audits by credit reporting agencies could be just as effective as OPC audits, as there is a genuine industry interest in maintaining data accuracy. A typical audit could include a step where a selection of files was sent to consumers, as they will be in the best position to check the accuracy of the data.[15]
• Systemic issues
  It is important to remedy systemic accuracy issues. It is not clear that this has always been done effectively in the past, although there are some examples of large-scale data cleansing following complaints against particular credit providers. Managing systemic issues is discussed in more detail in this Report at Section 2.9 (page 18).
• Minimisation and simplification of data fields
  It is possible that data accuracy could be improved by minimising and simplifying the amount and type of data that is collected by credit reporting agencies. The Office of the Privacy Commissioner, for example, has expressed concern that ‘expanding the volume of information reported to credit reporting agencies has the potential to increase the level of inaccuracy’.[16] Links between the amount of data and data accuracy are discussed in more detail in this Report at Section 2.13 (page 23).

As data accuracy is one of the most significant issues in credit reporting it should ideally be dealt with in the proposed Privacy (Credit Reporting Information) Regulations. Data accuracy is not simply an operational issue – it is actually an essential compliance issue with significant consumer harm and human rights consequences.

However, the ALRC appears to be uncertain about the best location for the regulation of data accuracy:

Where specific concerns about data quality are serious and well-defined, and the solution is reasonably clear, it may be appropriate to deal with them through specific provisions of the Privacy (Credit Reporting Information) Regulations. In other cases, matters may be dealt with more effectively through detailed data quality requirements in the proposed credit reporting industry code, subject to the overriding obligation to ensure that credit reporting information is accurate, up-to-date, complete and not misleading.[17]

The ALRC appears to be suggesting an unusual regulatory arrangement – where the issue is simple and the solution is clear the requirements can be set out in the Regulations, but where the issue is complex and the solution is unclear it should be dealt with by a potential industry Code.

A solution adopted in this Report is that core data accuracy requirements should be located in the proposed Privacy (Credit Reporting Information) Regulations. Supplementary industry rules about data consistency (e.g. the rules and processes for the consistent presentation of data across a diverse industry) may be addressed in a potential industry Code.

---

[8] Catalyst market research for Veda Advantage, November 2005.

[9] Veda Advantage, Submission to the Australian Law Reform Commission Issues Paper 32 – Credit Reporting, March 2007, page 22, <https://www.vedaadvantage.com/doc_library/63/VedaAdvantage_ALRC_IP32_Submission_March2007.pdf>.

[10] Refer, for example, to Submissions to ALRC IP32 from the Consumer Credit Legal Centre NSW Inc. and the Consumers’ Federation of Australia.

[11] For example, the removal of over 65,000 records following concerns over the accuracy of default information submitted by One.Tel.

[12] OPC audits are described in some details in ALRC IP32.

[13] CHOICE – The Australian Consumers’ Association, Reporting on the Credit Reporters, Consuming Interest, Autumn 2004.

[14] Refer, for example, to Australian Privacy Foundation, Review of Privacy – Credit Reporting Provisions Issues Paper 32 – Submission to the Australian Law Reform Commission, March 2007, page 19, <http://www.privacy.org.au/Papers/CrRpting-ALRC-0703.pdf>; and Banking and Financial Services Ombudsman, Review of Privacy – Credit Reporting Provisions: Issues Paper 32 – Submission by Banking and Financial Services Ombudsman Limited, March 2007, page 16, <http://www.bfso.org.au/abioweb/ABIOWebSite.nsf/3f51d54074f36f08ca256bce00094be3/15f5fb12141475a3ca2572ba0010bdd6?OpenDocument>.

[15] Note that there is some industry concern that the Privacy Act 1988 (Cth) itself presents obstacles to using consumer records for quality audit purposes and contacting consumer in this way could be difficult. The proposed regulations may be able to clarify this issue and make it clear that clients can be contacted for the purpose of checking data accuracy.

[16] Office of the Privacy Commissioner, Submission to the Australian Law Reform Commission’s Review of Privacy – Issues Paper 32 Credit Reporting Provisions, 13 April 2007, page 42, <http://www.privacy.gov.au/publications/submissions/sub-alrc-ip32-credit-reporting-200704.pdf>.

[17] Paragraph 54.4, DP72.

[« previous] [next »] [up] [title page] [full contents]

[view all]

[up to research]