Conduct of the review

1.1 This is the report of the Inspector-General of Taxation's (IGT) review into the Australian Taxation Office's (ATO) use of data matching. It is produced pursuant to section 10 of the Inspector-General of Taxation Act 2003 (IGT Act 2003).

1.2 The review is one of three concurrent reviews examining aspects of the ATO's compliance approaches to individual taxpayers. The other reviews examine delay in refund processing arising out of the ATO's Income Tax Refund Integrity Program and its administration of the superannuation excess contributions tax.

1.3 This review was commenced pursuant to subsection 8(1) of the IGT Act 2003, following concerns raised with the IGT during consultation to develop his 2012-13 forward work program. In particular, stakeholders were concerned that the data used were inaccurate or incomplete and the ATO was not taking action to refine the data before seeking to compare it to information provided in individual income tax returns.

1.4 Moreover, submissions outlined concerns that the ATO appeared to be relying solely on the data, without further research, to take compliance action such as amending taxpayers' assessments. The communication between the ATO and taxpayers and their agents was claimed to be insufficient to enable taxpayers to understand and respond to the case against them. It was submitted that this adversely impacted both taxpayers and tax agents. At the very least, it has resulted in additional time and costs in challenging ATO decisions.

1.5 Terms of reference for this review were announced on 20 November 2012. A copy of the terms and the submission guidelines are reproduced in Appendix 1. In response, the IGT received submissions from a diverse stakeholder group including taxpayers, tax agents and their respective representative bodies as well as data management specialists. The IGT also met with interested stakeholders to better understand their experience or those of their clients in dealing with the ATO on this matter.

1.6 Stakeholders expressed particular dissatisfaction with the ATO's use of data matching in two areas, namely, the administration of Capital Gains Tax (CGT) and Foreign Source Income (FSI). Therefore, whilst, this review examines the ATO's use of data matching at a general level, CGT and FSI have been used as illustrative examples. Taxpayer experience in these areas have been used to identify issues and opportunities for improvement which may be applied more broadly.

1.7 The IGT review team liaised with ATO staff in the Data Matching and Compliance Strategies (DMCS) stream of the Micro Enterprises and Individuals (MEI) business line, who have primary responsibility for the management of the data matching program. In addition, the IGT examined case documents from the ATO's data matching case management system to understand taxpayer complaints in this area and analysed ATO statistics in relation to data matching across a number of different areas.

1.8 The IGT also worked progressively with ATO senior management to distil the areas for improvement and to agree on specific actions.

1.9 In accordance with section 25 of the IGT Act 2003, the Commissioner was provided with an opportunity to make submissions on any implied or actual criticisms in this report.

What is data matching?

1.10 Each year, the ATO receives approximately 12.4 million income tax lodgements from individual taxpayers.1 The ATO takes the approach that it is neither possible nor realistic to expect that all individual income tax returns be manually reviewed for correctness.

1.11 The ATO uses a range of techniques, including data matching, to assist in its verification of income tax returns lodged by individual taxpayers. The ATO's data matching activities broadly consist of automated processes and analytics which compare taxpayer-reported information against information obtained by the ATO from third parties. These activities are supplemented by manual intervention for further investigation or clarification of any discrepancies.2

1.12 The ATO has stated that:3

Data matching is a powerful administrative and law enforcement tool. It allows information from a variety of sources to be brought together, compiled and applied to a range of public policy purposes at vastly lower costs than manual methods.

In revenue collection agencies such as ours, it helps us to identify people who are not complying with their obligations, and to detect fraud on the Commonwealth.

1.13 It is useful to note that data matching is designed to enable the ATO to detect instances of omitted income in individual income tax returns. This operates in tandem with the ATO's Income Tax Refund Integrity Program4 and its High Risk Refunds5 program which utilise analytical models and expert business rules to detect instances of over-stated or potentially fraudulent claims for deductions, offsets and other credits.

1.14 An important difference between data matching and the refund analytical models is that the former operates in the post-issue environment whereas the latter are applied pre-issue with one exception. The exception is the pre-filling service, for individual income tax returns, which operates pre-issue and which the ATO considers to have assisted in reducing the number of discrepancies identified post-issue.

1.15 Pre-issue compliance activity involves the ATO taking action to verify the details and amounts reported in an income tax return before a notice of assessment issues to the taxpayer. Post-issue compliance activity is compliance verification action taken after a notice of assessment has been issued to the taxpayer.

History and development of ATO data matching

1.16 Data matching has been a part of Australian tax administration for many decades. In its earliest years, at the end of the 1930s and early 1940s, the ATO manually matched interest reported by banks against interest reported in individual income tax returns.

1.17 With advancements in technology and the introduction of self assessment, including the enactment of Tax File Number (TFN) legislation by Parliament in the 1970s and 1980s, the ATO has moved towards an automated system of data matching through its development and implementation of the Information Matching System (IMS).6

1.18 With progressive updates and technological developments, IMS evolved into the current suite of systems used by the ATO, generally referred to as the ATO matching system (ATOMS). ATOMS comprises a suite of other systems including the Information Matching Analysis and Selection (IMAS) system, Automated Document Dispatch (ADD), Compliance Online Enquiry and Amendment System (COEAS) and the Windows Case Actioning System (WinCAS). These systems perform different functions within the ATO's data matching program. Their functions may broadly be described as follows:7

  • IMAS is the primary system for identifying discrepancies in legislative data reported by third parties, enabling the ATO to apply a range of business rules before selecting cases for action in ATOMS and uploading cases in to WinCAS for letter generation;
  • WinCAS is the primary case actioning system for data matching cases and is also the repository of notes and other details regarding the actioning of cases, enabling the ATO to automatically finalise a case with no further action, progress automated amendments in appropriate cases and allow ATO officers to reverse any amendments (or adjustments); and
  • COEAS provides an interface for ATO officers to view or edit legislative data provided by third parties, the details and amounts reported by taxpayers and any discrepancies between the two.

1.19 The ATO's enterprise data warehouse is the primary platform used to store all non-legislative data reported by third parties. From this system, the ATO applies a range of business rules to the data before selecting cases for audit in ATOMS and letters issued to taxpayers.8

1.20 In addition to the above systems, the ATO also utilises a 'Warning network' within WinCAS to generate advisory letters to taxpayers where data matching identifies small discrepancies which do not warrant compliance action. The ATO notes that such letters:9

... are not audit cases and will not amend. If a taxpayer remains discrepant in the following year an audit case may generate irrespective of the discrepancy amount.

1.21 The ATO explains that where this occurs, a note will appear on the WinCAS system to inform ATO officers that an advisory letter had issued to the taxpayer and the case was being progressed to audit due to the taxpayer's continuing non-compliance.10

Scope of the ATO's data matching activities

1.22 Data matching forms a significant part of the ATO's annual compliance program each year.11 The ATO's data matching activities may be broadly grouped into two categories. These are:

  • Business As Usual (BAU) data matching is an automated process in which individual income tax returns are matched against data obtained under legislation from such sources as employers and financial institutions to detect cases of potential underreported income or to verify taxpayers' entitlement to offsets or obligations such as the medicare levy surcharge; and
  • specific, or special purpose, data matching projects in which different ATO business lines acquire and match external data with information on ATO databases to identify and address risks in certain areas.12

1.23 In recent years, with improved automated processes and systems, the volume of the ATO's data matching transactions has increased significantly. This growth is shown in Table 1 below.

Table 1: ATO data matching transactions
  2007-08 2008-09 2009-10 2010-11 2011-12
Transactions matched 408m 502m 516m 539m 649m

Source: Commissioner of Taxation Annual Report 2011-12

1.24 In relation to this growth, the ATO further explains that:13

In 2011-12, we completed almost 600,000 audits and reviews resulting in adjustments. Over 97 per cent of these related to the micro enterprises and individuals markets, which included work in data matching, omitted income from third party data, employer obligations, and income tax refund data.

1.25 As the ATO's capability in data matching continues to develop, the scope of related activities, specifically special purpose data matching, has continued to expand, addressing an ever-widening range of areas and risks. A sample of special purpose data matching projects which were published in the Commonwealth Gazette in recent months includes projects targeting:14

  • income derived from online sales;
  • employers' lodgement and payment obligations;
  • claims of dependent tax offsets; and
  • contractor services to local councils and governments.

1.26 A complete list of the special purpose data matching projects which the ATO has announced or has completed in the past three financial years is contained in Appendix 3.

ATO resourcing and budget for data matching activities

1.27 Data matching for the individual taxpayer market segment is managed by ATO staff within the DMCS stream of the MEI business line. Over the past four years, the ATO's budgeting for full time equivalent (FTE) staff in this area has increased, as has the expected outputs for cases to be actioned and taxation revenue to be raised. A summary of this is outlined in Table 2.

Table 2: DMCS budget, staffing and work projections
  2009-10 2010-11 2011-12 2012-13
Budget ($) 10,641,365 17,034,164 19,400,000 29,783,000
Budgeted FTE 135 196.36 234.09 344.8
Budgeted FTE for audits 128.9 151 173.6 272.30
% of budgeted FTE for audits 95.5% 76.9% 74.1% 78.9%
Number of cases to be actioned 653,000 500,500 690,000 595,000
Amount of revenue to be raised ($) 253,000,000 331,000,000 838,000,000 995,985,000

Source: DMCS stream plans 2009-10, 2010-11, 2011-12, 2012-13

1.28 It is important to appreciate that not all budgeted FTE staff are allocated to the management of audits and audits arising out of data matching. In reviewing the DMCS's stream plans for the 2009 to 2013 financial years (inclusive), the IGT has extracted and included the number of budgeted FTE staff expected to undertake audits.

1.29 It is also interesting to note that for the 2012-13 year, the ATO expects to action a lower number of cases than in the previous year but raise a higher level of revenue. This suggests that the data matching projects in 2012-13 will focus on risks relating to higher revenue impacts.

ATO reallocation of staff

1.30 The ATO has advised that in December 2011, the data matching section was asked to undertake projects in addition to its usual data matching program. This work necessitated the injection of 160 FTE staff who were due to begin work in early February 2012. A large portion of these new staff members had recently completed the ATO's Graduate Program and had not previously undertaken data matching work.15

1.31 The ATO acknowledges that the short lead time between announcement of the staffing changes and the data matching projects on which they were to work represented significant challenges in terms of training and development as well as accommodation issues. In addition, it was noted that some of the new staff were dissatisfied with the announced change. These new officers were allocated to CGT and the Offshore Voluntary Disclosure Initiative (OVDI) in FSI. The ATO concedes that this resulted in some of the negative feedback from taxpayers in these projects.16

1.32 In 2012, the ATO has advised that data matching continued to employ new staff, including those from the ATO Graduate Program. However, the ATO has advised that with the benefit of hindsight and learnings from the previous year, the DMCS management team was better placed to engage with incoming staff, consider issues of resource allocation and accommodation in a more timely manner.17 The Capability Support and Assurance team was to ensure a consistent approach in creating and maintaining procedures, skilling, quality assurance and stream communications. Additional resources were reassigned to refine existing training packages for staff that were temporarily allocated to the DMCS and to provide additional support to new case actioning teams.18

1.33 The ATO has advised that no significant short term staff injections are expected in 2013-14.

1.34 In addition to the ATO's existing business budget, the government announced as part of its Mid-Year Economic and Fiscal Outlook (MYEFO) for 2011-12, that it would provide additional funding for the ATO to increase its data matching resources through to 2014-15,19 as outlined in Table 3 below.

Table 3: Additional government funding for the ATO's data matching resources
Revenue ($m)
  2010-11 2011-12 2012-13 2013-14 2014-15
Australian Taxation Office - - 174.2 145.4 116.7
Related expense ($m)
Australian Taxation Office
- - 17.0 20.4 20.3
Related capital ($m)
Australian Taxation Office
- - - 0.6 -

Source: MYEFO 2011-12

1.35 Further additional funding was announced in the MYEFO for 2012-13 to assist the ATO in expanding its data matching activities with a focus being the identification of lost superannuation.20

1.36 In the 2013-14 Federal Budget, the government announced further funding to assist the ATO to enhance its data matching work. The additional funding, expected revenue and cash receipts are outlined in Table 4.

Table 4: Further government funding for new reporting systems
Revenue ($m)
  2012-13 2013-14 2014-15 2015-16 2016-17
Australian Taxation Office - - 115.1 242.7 252.4
Related expense ($m)
Australian Taxation Office
- 8.8 20.2 24.2 22.3
Related capital ($m)
Australian Taxation Office
- 1.6 0.6 - -

Source: Budget 2013-14, Budget Paper No. 2

1.37 The government expects that the ATO would deliver $610.2 million in additional revenue over three years from 2014-15 to 2016-17 (inclusive). In underlying cash terms, the ATO is expected to deliver $431.7 million in receipts.21

1.38 As part of the Budget announcement, in addition to improving pre-filling, the government noted:22

The measure will establish new and strengthen existing reporting systems for:

  • taxable government grants and specified other government payments;
  • sales of real property, shares (including options and warrants), and units in managed funds;
  • sales through merchant debit and credit services;
  • managed investment trust and partnership distributions, company dividend and interest payments; and
  • transactions reported to the ATO by the Australian Transaction Reports and Analysis Centre.

1.39 In practical terms, it is expected that changes will be made to the current data reporting regimes in two broad areas. Namely, these will involve:23

  • aligning annual interest and investment reports with managed fund standard distribution statements, expanding information captured through the Australian Transaction Reports and Analysis Centre (AUSTRAC) as well as improving reporting of partnership distribution, company interest and dividend payment; and
  • introducing new reporting systems for taxable government payments and grants, sale of property, sale of shares and units in managed funds and sales histories reporting through merchant credit and debit card facilities.

1.40 The ATO has advised that any proposed legislative changes to effect the above measures are likely to commence operation from 1 July 2014 subject to relevant industry consultation.24

1.41 As detailed in the following chapters of this report, the IGT observes that the above measures may address some of the concerns raised through submissions in this review. Specifically, the development of new reporting systems and the strengthening of existing ones in relation to real property, share disposals and AUSTRAC should result in more timely and accurate data being provided to the ATO.

Data collection

1.42 To facilitate its data matching projects, the ATO relies upon information provided by third parties.

1.43 The ATO receives data from third parties under specific legislative obligations (legislative data) and other arrangements such as a memorandum of understanding or a demand under the access provisions available to the ATO under certain tax legislation (non-legislative or special purpose data). In the case of the latter, the access provisions do not empower the ATO to request any information not already in existence or to compel information providers to comply with specific formatting. As such, the application of such information in the ATO's non-legislative data matching projects may vary due to quality and integrity issues.

1.44 It should be noted that the ATO's exercise of its general access powers is not without controversy. As illustrated in Australia and New Zealand Banking Group Limited v Konza25 and its subsequent Full Federal Court appeal,26 concerns may arise for taxpayers in relation to both the subject and scope of the Commissioner's information requests.27 In particular, it was argued in this case that complying with the notices for information from the Commissioner would have resulted in ANZ breaching common law and statutory secrecy and confidentiality requirements within Vanuatu.28 Moreover, arguments were also raised regarding the breadth of the notices, applying to all customers who have or have had accounts with ANZ Vanuatu.29

1.45 The ATO has advised the IGT that the above matter is unique for a number of reasons. Firstly, the data collection was initiated by the Large Business and International business line and not the DMCS. Secondly, the use of compulsory access powers is not often required within the DMCS context unless the data provider specifically requests that the ATO issue a formal notice. Thirdly, through negotiations between the DMCS and third party data providers, the ATO notes that the levels of dispute and litigation highlighted above do not generally occur.

1.46 Acknowledging the ATO's advice in relation to this case, and noting that the Court dismissed the taxpayer's application and appeal in this matter, the case nonetheless serves to highlight concerns which may arise in the course of the ATO's data collection activities.

1.47 It is also important to recognise that the provision of data to the ATO, both legislative and non-legislative, imposes a compliance cost on these information providers. In respect of legislative data, costs manifest in a number of ways, including through implementation of specific systems to capture and report required information, formatting to meet the ATO's requirements or processes to meet legislative timeframes which may vary between industries and stakeholder groups. In relation to non-legislative data, the costs may include time and staffing resources required to extract, format and deliver the data.

1.48 To an extent, the ATO recognises the impost created for third party data providers, particularly those who provide data at the ATO's request rather than in accordance with business as usual taxation reporting requirements.30 The ATO' internal practice statement, in relation to non-legislative data, contemplates reimbursement of expenses incurred by third party providers in certain circumstances.31 However, the IGT notes that such reimbursements are wholly discretionary and the practice statement does not outline any specific considerations or factors which would lead to reimbursement being provided to the third party provider.

1.49 The ATO has advised the IGT that the absence of specific directions and factors within the internal practice statement enables flexibility for it to consider each request for reimbursement on a case by case basis. It has further advised that under the leadership of the current Assistant Commissioner, the ATO has not refused any requests for reimbursement of extraction or formatting costs.32

Legislative data

1.50 Legislative data is collected under specific legislative reporting requirements.33 The ATO notes that:34

This [data] forms the core of the ATO's corporate data holdings and includes return forms and schedules, Business Activity Statements, Payment Summaries including welfare payment details and annual investment income details provided by investment bodies.

1.51 When and how the data is provided is prescribed by legislation and the ATO is able to specify the fields and formats that it requires of the data being provided. In addition, such data is subject to TFN quotation rules which add an additional layer of quality and integrity. The TFN as an identifier also enables the ATO to readily match, with a high level of certainty, the third party data against information in the income tax returns lodged by taxpayers.

1.52 The ATO, therefore, considers that the quality of legislative data is extremely high.35

Non-legislative data

1.53 Non-legislative data is provided to the ATO in response to a request under its general access powers,36 a memorandum of understanding between the ATO and the data provider or through formal information exchange programs, such as those the ATO maintains with foreign revenue authorities. Non-legislative data is acquired for a special purpose.

1.54 As such data is not provided under a specific legislative requirement, the ATO is reluctant to prescribe what data should be provided and how. Therefore, the quality and amount of the data is limited to that which has been collected and stored for the data provider's business purposes which may not necessarily align with those of the ATO. This requires the ATO to work closely with the information providers to ensure that data available are suitable for the ATO's purposes.

1.55 The ATO notes that the content and format of non-legislative data varies as between providers and industries. The ATO also acknowledges the varying quality and levels of completeness of such data.37

1.56 In addition, and unlike legislative data, there are often no unique identifiers such as dates of birth, TFNs or Australian Business Numbers (ABN), thereby limiting the effectiveness of identity matches between the data and relevant taxpayers.

1.57 As reflected in the earlier discussion on government funding, the limitations on non-legislative data are well-recognised. In addition to providing more funding for enhancing the reporting of this type of data, the government is to also consult with key stakeholders, including the States and Territories, on the design of any new data gathering and reporting systems focusing on taxable government payments and grants, sale of property, sale of shares and units and sales histories from merchant credit and debit card facilities.38

Capital gains tax and foreign source income data matching

1.58 As already indicated, most submissions to this review focused on specific non-legislative data matching projects in two areas, namely, CGT for the disposal of real property or shares and FSI. A brief outline of each of these areas is provided below.

CGT data matching

1.59 CGT is tax paid on gains received following the disposal of a capital asset, such as real property (subject to certain exemptions) or shares owned by the taxpayer that is not otherwise income.39 Capital gains are not taxed separately. They form a part of a taxpayer's assessable income which needs to be reported annually.

1.60 There is a risk that taxpayers may not have declared capital gains received during the year and, consequently, not fully reported the extent of their assessable income.

1.61 To address this risk in relation to real property and share disposals, the ATO obtains data from a range of state-based revenue offices, land titles registries and share trading registries and compare these data to taxpayers' returns to identify whether any disposals of capital assets occurred in the financial year and, if so, whether the income derived from such disposals was reported.

FSI data matching

1.62 The ATO notes in relation to FSI cases that 'Australian residents are taxed on their worldwide income. This means they must declare all income received from foreign sources in their income tax returns.'40

1.63 The ATO has identified as a risk the potential for FSI to be undeclared or incorrectly treated by taxpayers as exempt.41

1.64 To address this risk, the ATO relies upon data such as those obtained from AUSTRAC,42 taxpayers' holdings in offshore bank accounts from financial institutions as well as financial information obtained from foreign revenue authorities through tax information exchange agreements, tax treaties and other information sharing arrangements.43 These data are compared to taxpayers' returns to determine whether there may be any income received by taxpayers from foreign jurisdictions which need to be declared.

ATO Data Management

How the ATO manages legislative data

1.65 Legislative data forms part of the ATO's corporate data holdings. As such, it is subject to privacy and secrecy rules contained in the relevant taxation legislation as well as other information management legislation which binds the ATO. Such legislation includes requirements under the Privacy Act 1988 and Division 355 of Schedule 1 to the Taxation Administration Act 1953 (the TAA 1953).

1.66 In addition, the ATO has also developed a number of internal policies and controls in relation to maintaining and protecting confidential data and limiting access to such data only to staff with appropriate security clearance levels who require access in the course of their duties. ATO officers who have such access must personally keep that data confidential and must not use it for any other purpose.

How the ATO manages non-legislative or special purpose data

1.67 Non-legislative or special purpose data is collected by the ATO at the request of specific business lines to address potential risks within certain ATO products or market segments. The ATO's processes for collection and management of this data are governed by its internal Corporate Management Practice Statement CMPS 2004/17 Special Purpose Acquisition Data.

1.68 Specifically, paragraphs 11 to 13 of the updated CMPS 2004/17 states:44

11. The ATO's policy on Data matching special purpose acquisition data is to ensure the acquisition and use of data complies with our legislative and other obligations, as detailed in the following points:

  • The ATO voluntarily complies with the Office of the Australian Information Commissioner's The use of data matching in Commonwealth administration - Guidelines (February 1998). These Guidelines provide direction to ATO personnel for the acquisition, use, retention and destruction of data as well as reporting obligations.
  • ATO personnel must adhere to taxation and other Commonwealth legislation that is for the protection of information obtained by ATO personnel in the course of their duties. This includes the Information Privacy Principles (IPPs) in section 14 of the Privacy Act 1988 that cover the collection, storage, use and disclosure of personal information.

12. The Special Purpose Data Steward must be consulted for all proposed collections of special purpose acquisition data and the Assistant Commissioner Data Matching Compliance Strategies must approve the data matching strategy before any contact is made with a data provider.

13. The ATO does not pay for data where the Commissioner's formal access powers have been exercised. However, in certain specific circumstances the ATO may reimburse data providers for expenses incurred in providing data in a format suitable for use by ATO systems.

1.69 The ATO has prescribed specific roles at a senior executive level both within the MEI and other business lines in relation to processes to be undertaken before special purpose data is requested, how it is managed in the ATO's data warehouse and who is responsible for matching the data against the ATO's identification system before applying it to taxpayer records.45 A list of key roles and responsibilities in this process is contained in Appendix 4.

1.70 Generally, the process begins where an ATO business line identifies a need for data to assist in its active compliance work. Once a risk is identified, the DMCS stream requires business lines to consult with the ATO's Data Matching Steward to discuss the underlying risk sought to be treated and the proposed data to be used.46 The Data Matching Steward position commenced operation within the ATO from 1 July 2013 and their role is to guide and assist the business line to explore the possibilities of using third party data to address the identified risk and potential sources of data which may be available.

1.71 The Data Matching Steward specifically assists the business officer to prepare a brief for endorsement by a senior executive within the business line and formal approval by the Assistant Commissioner DMCS.47 The brief outlines the compliance risk, strategy for acquiring and using the data and associated risks.

1.72 In addition to the Data Matching Steward, the ATO has advised the IGT that each business line has nominated information gathering advisors who can also assist and provide advice on different aspects of data access, validation and application to different risks.

1.73 Once the Data Matching Steward and the business line have reached a common understanding, the former refers the business line to an appropriate senior officer within the DMCS. The DMCS officer undertakes negotiations with the intended data supplier to discuss issues such as data format, quality, completeness, strengths, weaknesses as well as any industry issues and intelligence.48 The DMCS officer obtains the data, either as a pilot or a complete set and undertakes a process to analyse it within the ATO's Data Warehouse.

1.74 The DMCS officer then considers and discusses the development of appropriate protocols in compliance with the Data-Matching Program (Assistance and Tax) Act 1990 (the DMPAT Act 1990) Guidelines49 or the Guidelines for the use of Data Matching in Commonwealth Administration50 issued by the Australian Information Commissioner (AIC). The former are guidelines which bind all data matching projects under the DMPAT Act 1990 while the latter are voluntary guidelines for data matching projects not conducted under the DMPAT Act 1990. The ATO considers both sets of guidelines in the conduct of its data matching projects.51 These discussions are conducted between the DMCS officer, the Data Matching Steward and the business line representative.

1.75 Upon agreement of the approach and protocols, the business line completes the appropriate documentation which is referred to the relevant Deputy Commissioner for approval. Once these documents are approved and provided to the AIC, the DMCS makes the data available to the business line.52

1.76 The business line uses the data and reports on the usage, outcomes and governance to the DMCS and Data Matching Steward. The DMCS and the Data Matching Steward use this information to report to the AIC on a periodic basis, or as requested.53

1.77 While it is not presently reflected in the assurance process, the ATO has provided information to the IGT in relation to its pilot testing of new risks sought to be treated through data matching. Reports of the pilots assess the effectiveness of data matching in these areas and make recommendations as to whether a broader application is appropriate.54

1.78 The ATO's broad assurance process for special purpose data is illustrated Figure 1 below.

Figure 1: Micro Enterprises and Individuals data matching assurance process

Chart describing the Micro Enterprises and Individuals data matching assurance process.

View image enlarged

Source: adapted from ATO information

Reviews by the Australian National Audit Office

1.79 The ATO's use of data matching and analytics generally and its use of data matching in CGT compliance work have been the subject of separate reviews by the Australian National Audit Office (ANAO).

1.80 In its report, Administration of Capital Gains Tax Compliance in the Individual Market Segments,55 the ANAO examined three key areas, namely, governance, identification and assessment of compliance risks and compliance activities with a focus on real property and share disposals.56

1.81 Amongst other things, the ANAO's report observed the difficulties associated with the variation in data sets obtained from different state revenue offices and other state-based sources and the challenges the ATO faced in converting these into a useable format.57 The ANAO made a number of recommendations, including the need for the ATO to provide feedback to state revenue offices on errors or defects in the data provided. An assessment was also made as to whether 'automated data matching and case actioning is the most appropriate strategy to achieve capital gains tax compliance in the individuals market segment.'58

1.82 The ATO agreed with all of the recommendations made by the ANAO.

1.83 The ATO's use of data matching and analytics was explored more generally in the ANAO's 2008 report, The Australian Taxation Office's Use of Data Matching and Analytics in Tax Administration.59

1.84 In that review, the ANAO examined certain key themes including the ATO's 'strategic goals and governance arrangements for data matching and analytics, its compliance with privacy requirements and whether the [ATO] is achieving intended results, which include revenue collection, optimised compliance and provision of improved services to taxpayers.'60

1.85 As a general observation, the ANAO found that:61

The Tax Office is making use of its data matching and analytical capabilities in a more corporate and strategic way. This has contributed to the Tax Office reporting improved compliance, better services and the more efficient and effective use of resources. It has also enabled the Tax Office to better understand risks.

1.86 The ANAO made six recommendations which, in the main, were aimed at enhancing data matching in relation to all taxpayers rather than just individuals, increasing the use of pre-filling and improving corporate efficiencies through clarification of roles and responsibilities and the development of a corporate plan for the collection and use of external data. The ATO agreed with all recommendations made by the ANAO.62

Stakeholder concerns regarding the ATO's use of data matching

1.87 Submissions received by the IGT have indicated that stakeholders generally do not take issue with the ATO's use of data matching in its compliance activities, particularly in relation to pre-filling of income tax returns. However, these submissions also raised some concerns regarding the ATO's processes generally and action taken once a potential discrepancy is identified. Broadly, these concerns include:

  • a perception that the data used by the ATO is not reliable or current and the ATO is not taking sufficient action to refine the data before applying it to matching models;
  • the ATO's data matching processes are not timely and the associated costs are disproportionate to the level of adjustments and level of revenue protected;
  • the ATO is taking inappropriate action based on the data, such as issuing default assessments without providing taxpayers with sufficient opportunity to address its concerns; and
  • the ATO is not properly communicating and engaging with taxpayers to ensure they understand the ATO's concerns and the reasons for any identified discrepancies.

Impact of data matching on taxpayers and tax agents

1.88 While a number of taxpayers and tax agents have indicated their general support for the ATO's use of data matching, particularly in relation to assisting taxpayers to comply with their obligations through pre-filling of income tax returns, stakeholders have also submitted to the IGT that in some cases data matching has adversely impacted both taxpayers and tax agents.

1.89 In the case of taxpayers, these impacts have been expressed as:

  • the lack of clarity by the ATO as to the purpose of data matching and nature of its inquiry has resulted in taxpayers taking unnecessary action;
  • some taxpayers are experiencing stress and anxiety at what they perceive to be a heavy-handed approach by the ATO and the feeling that the ATO considers them to be fraudulent or incompetent in the management of their tax affairs despite good compliance histories; and
  • in some cases, taxpayers accepted and paid the adjusted assessments without challenge as they perceived that the time and cost associated with challenging the adjustment would exceed the adjustment itself.

1.90 For tax agents, these impacts were expressed as the additional time and cost expended to deal with the ATO. These, in turn, have created difficulties for the tax agent who may be absorbing the cost of the dealing with the ATO or creating tensions with their clients who do not always accept that action taken by the agent was necessary.

1.91 Further tensions between tax agents and their clients may also arise where the ATO contacts taxpayers directly as this may lead to stress and anxiety on the part of the taxpayer. This has a flow on effect to the tax agent who must intervene and seek to assuage the taxpayer's concerns before responding and addressing the ATO's enquiries.

1.92 Similar impacts were submitted to the IGT in relation to his review into the ATO's Income Tax Refund Integrity Program.63 As in that report, the IGT considers that the ATO should consult with the tax agent community and implement strategies to ensure that specific data matching correspondence is directed to the person designated by the taxpayer. The ATO has advised the IGT that its current processes are aimed at achieving this outcome, that is, correspondence is directed to the address on the taxpayer's most recent income tax return or as separately advised to the ATO. A further action that the ATO could explore is informing tax agents of data matching correspondence which may issue to their clients and on which their clients may be seeking advice or assistance.

1 Australian Taxation Office (ATO), Compliance Program 2012-13 (2012), p. 10.

2 Ibid, pp 19-20; Office of the Privacy Commissioner, The Use of Data Matching in Commonwealth Administration -Guidelines, February 1998, p. 2.

3 ATO, Data matching - overview of our compliance approach (5 September 2012).

4 Inspector-General of Taxation (IGT), Review into the Compliance Approach to Individual Taxpayers - Income Tax Refund Integrity Program, September 2013.

5 Australian National Audit Office (ANAO), Administration of High Risk Refunds in the Individuals and Micro Enterprises Market Segments, Audit Report 12, 2007-08. Since 1 July 2013, the ATO has integrated all aspects of the High Risk Refunds program which relates to individual income tax risks into its Income Tax Refund Integrity Program.

6 ATO, Working for All Australians 1910-2010 (2010) p. 147.

7 ATO, 'Systems Actions', internal ATO document.

8 ATO, communication with the IGT, 30 August 2013.

9 Above n. 7.

10 Ibid.

11 ATO, Compliance Program 2010-11 (2011) pp. 5, 19, 21; ATO, Compliance Program 2011-12 (2012), pp. 3, 15, 18; ATO, Compliance Program 2012-13 (2013) pp. 10 and 27.

12 Practice Statement CM PS 2004/17 Data matching - special purpose acquisition data, internal ATO document.

13 Commissioner of Taxation, Annual Report 2011-12, p. 94.

14 See for example: Commonwealth, Government Notices Gazette, GN4 2013 (30 January 2013); Commonwealth, Government Notices Gazette, GN 12 2013 (27 March 2013) and Commonwealth, Government Notices Gazette, GN 13 2013 (3 April 2013).

15 ATO, communication with the IGT, February 2013.

16 ATO/IGT workshop, 18 February 2013.

17 Ibid.

18 Above n. 15.

19 Commonwealth, Mid-year Economic and Fiscal Outlook 2011-12 (29 November 2011) p. 176.

20 Commonwealth, Mid-year Economic and Fiscal Outlook 2012-13 (2012) p. 10.

21 Commonwealth, Federal Budget 2013-14, Budget Paper No. 2, p. 44.

22 Ibid.

23 ATO, communication with the IGT, 22 August 2013.

24 Above n. 22.

25 Australian and New Zealand Banking Group Limited v Konza [2012] FCA 196.

26 Australian and New Zealand Banking Group Limited v Konza [2012] FCAFC 127.

27 Above n. 25 at [3] and [4]; Ibid at [9].

28 Above n. 25 at [3].

29 Ibid at [4] and [5].

30 ATO, Access and Information Gathering Manual, para. 1.8.19 [this manual is subject to revision. This reference is current as at 18 June 2013].

31 Above n. 12, para. 13.

32 Above n. 22.

33 See for example regulation 56 of the Income Tax Regulations 1936 in relation to Annual Investment Income Reports.

34 Above n. 15.

35 Above n. 15.

36 For example, section 264 of the Income Tax Assessment Act 1936 for income tax and section 353-10 of Schedule 1 to the Taxation Administration Act 1953 for indirect taxes.

37 ATO, ATO Tax Practitioner Forum Minutes, August 2012, Item 5.

38 Above n. 21, p. 44; Above n. 22.

39 ATO, Capital Gains Tax, 28 June 2013.

40 ATO, Foreign Source Income case actioning procedures, 18 December 2012, internal ATO document.

41 Section 23AG of the Income Tax Assessment Act 1997.

42 Australian Transactions Reports and Analysis Centre.

43 ATO, International Tax Agreements, 7 January 2013.

44 Above n. 12, paras. 11 - 13.

45 Ibid.

46 Corporate Management Procedure and Instructions 2004/17/01, para. 6.

47 Above n. 46, para. 7.

48 Ibid., para. 7.

49 Issued pursuant to section 12 of the Data-Matching Program (Assistance and Tax) Act 1990.

50 Office of the Australian Information Commissioner, Guidelines for the use of Data Matching in Commonwealth Administration, February 1998.

51 See for example, Office of the Australian Information Commissioner, Annual Report 2011-12, pp. 66, 68-72.

52 Above n. 46, paras. 8-10.

53 Data-Matching Program (Assistance and Tax) Act 1990 Guidelines, section 8.

54 ATO, Shares final pilot report, 2013, internal ATO document; ATO, PTU final pilot report, 2013, internal ATO document.

55 ANAO, Administration of Capital Gains Tax Compliance in the Individual Market Segments, 20 December 2006.

56 Ibid., p. 39; In 2003-04, the ANAO also undertook a review into the Australian Taxation Office's Management and Use of Annual Investment Income Reports, which contain data relevant for the identification of share disposals.

57 Ibid., pp. 52-57.

58 Ibid., pp. 57 and 59.

59 ANAO, The Australian Taxation Office's Use of Data Matching and Analytics in Tax Administration, 24 April 2008.

60 Ibid., p. 16.

61 Ibid., p. 17.

62 Above n. 59, pp. 28 and 29.

63 Above n. 4.