Introduction

4.1 In this review, the IGT examined the ATO's management of the non-lodgement of individuals' personal income tax returns. The review was undertaken in response to stakeholder concerns that large amounts of non-lodged tax returns may pose a risk to the integrity of the tax administration system. The issues that the IGT identified in independent research were also a factor in deciding to conduct this review. Terms of reference for the review were released on 12 October 2007.

4.2 The review observed that every year between 1.2 and 1.5 million individual taxpayers failed to lodge their tax return by the required dates. However, it also noted that this taxpayer failure was not attributable to ATO mismanagement but rather a function of the self assessment system in which nearly all taxpayers are required to lodge annually and where the ATO's resourcing does not permit it to pursue all non-compliance. The IGT noted a key reason for the large accumulation of outstanding tax returns was that the Commissioner has limited powers to exempt individuals from lodging returns.

4.3 The review observed that while a significant number of returns are not lodged annually, these returns did not represent a high risk to revenue or the integrity of the tax system. Furthermore, these non-lodging taxpayers were of a low risk, based on their profile and the ATO had strategies to identify taxpayers of higher risk.

4.4 Accordingly, the IGT's review found that, at the time, the ATO was managing the non-lodgement of individual returns well and with sophistication within the legislative framework. Notwithstanding this finding, the accumulation of outstanding low-risk tax returns can create an incorrect impression that a risk exists that needs to be addressed. The independent research, conducted at the time of the review, also found that 'the community expects the [ATO] to maintain individual non-lodgement at a slightly better level' than it was achieving at the time.49

4.5 The IGT's review identified a number of areas for ATO improvement, including the need for the ATO to enhance its public reporting on the levels of non-lodgement, clearer identification of high and low risk non-lodgers on ATO systems and increasing the use of default assessments in appropriate circumstances. The IGT also considered that the findings of the review and the research could usefully be considered in the Australia's Future Tax System review, which had sought input on income tax return simplification, including the removal of the requirement for certain taxpayers to lodge. The IGT also made recommendation for the previous Government to consider whether support for the ATO's use of third party data to identify non-lodgers should be increased and whether the failure-to-lodge penalty regime, particularly for high-risk taxpayers, should be strengthened.

4.6 In all, the IGT made six recommendations, three of which were directed to the ATO. The ATO agreed with all three recommendations. The remaining three recommendations, key recommendations 1, 3 and 5, were made for the previous Government to consider.

4.7 The then Minister publicly released the report for this review on 16 October 2009.

Implementation of agreed recommendations

4.8 The following section sets out the agreed recommendations and examines the ATO's implementation to the extent that they were agreed.

Key Recommendation 1

4.9 This recommendation was made for the previous Government to consider and is reproduced in Appendix 2.

Key Recommendation 2

The Inspector-General recommends that the Tax Office supplement its current reporting on lodgement compliance by a periodic report on the broader outcomes and impacts being achieved on the level of non-lodgement in the community.

ATO position — implemented

4.10 The ATO agreed with the recommendation, noting that the Commissioner would create an income tax lodgement report each financial year. The report would draw from existing data sources and other ATO research to provide an analysis of 'lodgement on time', 'late lodgement' and 'not lodged returns'.

4.11 Accordingly, the ATO adopted a number of initiatives to supplement its reporting on the level of non-lodgement in the community, the impacts of non-lodgement and the outcomes achieved through the ATO's compliance work in this area. Specifically, these initiatives are the:

  • income tax lodgement annual overview;
  • enterprise reporting program of work for trusted reporting;
  • targeted risk report; and
  • use of external consultancy reports; and
  • community perception surveys.

Income Tax Lodgement Annual Overview

4.12 In December 2010, the ATO produced its first Income Tax Lodgement Annual Overview report (the 2010 Report). This internal report provided an overall picture of the lodgement trends between 2004-05 and 2008-09 (inclusive) by drawing on statistical trend data in lodgement activity from a number of different sources. The 2010 Report also provided a broad overview of the ATO's compliance work to address non-lodgement.50

4.13 The general observation arising out of the 2010 Report was that there appeared to be an upward trend in lodgements than those required to lodge and also in lodgements being made on time. The 2010 Report also measured the ATO's compliance strategies in relation to lodgement compliance across a number of taxpayer segments as well as supporting future directions to enhance lodgement compliance.

4.14 Since 2010, the ATO has included these statistics in the annual Income Tax Health of the System Assessment (HOTSA) process.51 The HOTSA process draws on a number of different resources to provide a structured mechanism to identify and assess risks in different areas of the ATO and make decisions regarding operational shifts to address these risks.

Enterprise Reporting Program of Work for Trusted Reporting

4.15 In early 2011, the ATO commenced the 'Enterprise Reporting Program of Work for Trusted Reporting'. One of the reports produced as part of this program was the Lodgment Performance Income Tax Report52 which is intended to provide a high level summary of two key measures for lodgement performance, being:

  • lodgements as a percentage of anticipated population; and
  • lodgements on time.

4.16 From July 2013, the ATO incorporated an Income Tax Return Not Required Analytical Model (the Model) into the enterprise lodgement reports to identify the expected or 'anticipated' population of taxpayers required to lodge income tax returns. This Model predicts those taxpayers not required to lodge and removes them from the statistical analysis of outstanding returns to enable a more accurate picture of non-lodgement risk.

4.17 The Model's use and outputs are discussed further below in relation to Key Recommendation 4.

Targeted Risk Reporting

4.18 As part of its risk management process, the ATO also undertakes specific targeted reporting which focuses on lodgement performance in the individuals' market segment for both income tax returns and business activity statements. The results of this monthly reporting are contained in the 'Individuals Segment Lodgment Performance Report', which includes information regarding:53

  • the number of active taxpayers and returns lodged;
  • comparative data for returns lodged on time and those lodged late using an age basis (for example, 28 days after due date and 90 days after due date); and
  • comparative data for returns lodged on time and lodged late using the percentage of returns that resulted in credit, debit or nil assessments.

4.19 This internal ATO report also presents data on different performance levels between self-preparers and those lodging through tax practitioners who may have access to extended lodgement timeframes. The ATO considered that such reporting provides better data regarding the categories of non-lodgers and the corresponding risks posed by their non-lodgement of returns.

External consultancy reports

4.20 In May 2010, the ATO engaged a consultancy firm to research the attitudinal and behavioural drivers of lodgement, examining such things as community perceptions on consequences for non-lodgement, the ATO's effectiveness in detecting non-lodgement and qualitative differences between taxpayers who lodged on time and those who did not.54

4.21 Another research study commissioned by the ATO found, amongst other things, that 72 per cent of survey participants believed they would be 'caught by the ATO' if they did not lodge their tax return. This percentage had decreased slightly from when a similar survey was conducted in 2006.55

4.22 The ATO has advised that at present no further research of this kind is planned.

Community Perceptions

4.23 The ATO also gauges community perceptions on a range of issues including perceptions on lodgement obligations56 as part of its Annual Research Program. However, the survey does not specifically seek to understand community perceptions on the ATO's work in relation to enforcing non-compliance in relation to lodgement obligations.

IGT conclusion — implemented

4.24 The IGT considers that the ATO has implemented this recommendation.

Key Recommendation 3

4.25 This recommendation was made for the previous Government to consider and is reproduced in Appendix 2.

Key Recommendation 4

The Tax Office should flag low-risk non-lodged returns in its systems and should identify them as a separate category in its management reports to enable a clearer focus on higher risk non-lodged returns.

ATO position — implemented

4.26 The ATO agreed with this recommendation.

4.27 In response to this recommendation, the ATO has made progress towards flagging and separating low-risk non-lodged returns so that focus may be directed towards higher risk non-lodged returns. In doing so, the ATO has:

  • continued the testing and refinement of the Model; and
  • implemented an Assessed Low Risk population table and an Assessed Low Risk accounting treatment indicator on the ATO's system.

Income Tax Return Not Required Analytical Model

4.28 In 2009, the ATO commenced the development of new reporting capability based upon implementation of the Model. As discussed earlier, the Model was developed to identify, amongst other things, whether a taxpayer was either required to lodge (R2L) or not required to lodge (NRQL)57 and enable the ATO to identify and exclude from the non-lodgement population those taxpayers who are not required to lodge. The aim was to provide a more accurate picture of individuals' non-lodgement and a clearer focus on higher risk non-lodged returns.

4.29 The ATO has advised the IGT that it has completed the development of the Model and has evaluated its effectiveness by comparing actual lodged overdue returns (between March 2012 and June 2012) against the original risk scores allocated to the particular taxpayer by the Model. In doing so, the evaluation sought to validate whether the Model had correctly predicted a taxpayer as either being required to lodge or not required to lodge.

4.30 The ATO evaluation also examined both strike rates and miss rates. Strike rates assessed the extent to which the Model correctly classified the taxpayer. Miss rates assessed the extent to which the Model misclassified a taxpayer. The ATO has provided statistics on the combined strike rates for R2L and NRQL for each of these taxpayer groups at the 'client' and 'form' levels. The client level strike rate indicates the effectiveness of the models to accurately determine whether a particular taxpayer is required to lodge in a given year. The form level strike rate indicates the effectiveness of the models to accurately identify whether certain income tax returns need to be lodged.58

4.31 The two types of strike rates are necessary and complimentary as a given taxpayer may be associated with multiple non-lodged income tax returns which may not be required to be lodged owing to the different capacities or roles the individual may be performing—for example, a partner in a partnership.59

4.32 Lastly, the evaluation examined both R2L and NRQL in relation to a range of taxpayer segments including individuals, companies, superannuation funds, taxable trusts, non-taxable trusts and partnerships. Moreover, it applied relevant risk designations based on the quantum of tax payable or refundable on the return. The ATO set a baseline benchmark of an 80 per cent strike rate and measured the Model's performance against this benchmark.

4.33 The ATO's statistics on the strike rates are set out in Table 4 and Table 5 below.60

Table 4: Required to Lodge (R2L) strike rates at the client and form levels
R2L Strike Rates at the Client and Form Levels (%)
Client Type Individuals Companies Superannuation Funds Taxable Trusts Non-Taxable Trusts Partnerships
Client Level 72 67 70 22 47 66
Form Level 70 62 71 22 43 59

Source: ATO

Table 5: Not Required to Lodge (NRQL) strike rates at the client and form levels
NRQL Strike Rates at the Client and Form Levels (%)
Client Type Individuals Companies Superannuation Funds Taxable Trusts Non-Taxable Trusts Partnerships
Client Level 78 87 97 92 92 87
Form Level 81 89 96 94 94 91

Source: ATO

4.34 In line with the above figures, the ATO's evaluation acknowledged that, while the R2L strike rate was better than random selection, some further improvements were warranted.61 Steps were taken to improve the Model's strike rate and, in July 2013, the Model was incorporated into the current reporting framework on low risk taxpayers. This framework is part of the ATO's annual HOTSA process62 and provides an enterprise-wide view of the lodgement population and associated lodgement risks.

Assessed Low Risk population table and Assessed Low Risk accounting treatment indicator63

4.35 In implementing this recommendation, the ATO initially used existing analytical models and business rules to identify an Assessed Low Risk population which was used to populate a table in the ATO's Enterprise Data Warehouse. This Assessed Low Risk population table enabled the Tax Practitioner and Lodgment Strategy (TPALS) business line to report on lodgement compliance for identified low risk taxpayers. This table was not used for reporting purposes.

4.36 Since March 2012, the ATO deployed an Assessed Low Risk accounting treatment indicator within the Integrated Core Processing system, which may be applied against a taxpayer's file for a given specified period. This indicator enabled automatic and manual identification of Assessed Low Risk taxpayers. These taxpayers were then excluded from compliance activity for the specified time period.

4.37 The ATO has advised, however, that the ability to manually invoke the Assessed Low Risk accounting treatment indicator required further testing and evaluation before being employed for more general operation. However, this further work was put on hold in February 2013 while the ATO focused on finalising the evaluation and deployment of the Model discussed earlier.

4.38 Following this evaluation the ATO concluded that the continued use of the Assessed Low Risk population table in the Data Warehouse was not warranted as the Model, combined with annual data matching, identifies and excludes low risk taxpayers from compliance activities with a sufficient degree of accuracy.

4.39 Where it is determined that the taxpayer does not have a requirement to lodge for a particular year, a Return Not Necessary (RNN) indicator is applied against the taxpayer's file. Taxpayers identified as having an RNN indicator are then excluded from compliance activities in relation to that return.

IGT conclusion — implemented

4.40 The IGT considers that the ATO has implemented this recommendation.

Key Recommendation 5

4.41 This recommendation was made for the previous Government to consider and is reproduced in Appendix 2.

Key Recommendation 6

The Tax Office should progressively increase, where appropriate, the use of default assessments to further support lodgement compliance.

ATO position — implemented

4.42 The ATO agreed with this recommendation and, in response, commenced its work in relation to issuing default assessments for:64

  • all micro enterprises, small to medium enterprises and individual taxpayers that fall under the Risk to Revenue project in 2010-11;
  • child support agency clients;
  • superannuation funds; and
  • cash economy businesses.

4.43 The ATO also refined and expanded the default assessment process following the IGT's review.65 The ATO uses this process as part of an end-to-end treatment strategy for persistent non-lodgers. The current process for considering and issuing such default assessments is outlined in Figure 1 below.

4.44 Relevantly, Figure 1 illustrates that the ATO will issue a default assessment warning letter where data exists to create a default assessment and the taxpayer has failed to respond to lodgement enforcement activity, such as ATO reminder letters and telephone enquiries. This letter sets out the amount of the default assessment and the applicable penalties.

Figure 1: Default assessment process map

Chart giving a default assessment process map. The ATO will issue a default assessment warning letter where data exists to create a default assessment and the taxpayer has failed to respond to lodgement enforcement activity, such as ATO reminder letters and telephone enquiries.

Source: ATO

View image enlarged

4.45 The ATO has advised that approximately 50 per cent of taxpayers who receive this warning letter respond by lodging their return. It also recognises that a proportion of taxpayers are not contactable.

4.46 The remaining group of taxpayers may be issued default assessments, depending on the resources available to manually create these default assessments and deal with subsequent taxpayer contact and objections. The ATO intends to automate this process and, therefore, envisages that it will be able to issue greater numbers of these assessments in future.

4.47 To illustrate trends in its use of default assessments, the ATO has provided the IGT with the numbers of default assessments issued between the 2008-09 financial year to 2012-13 year to date (YTD), as at 14 May 201366 across three taxpayer groups, being companies, individuals and superannuation funds. These are outlined in the Table 6 below.

Table 6: Number of default assessments issued between 2008-09 to 2012-13 YTD
Number of default assessment raised in various financial years
Client Type 2008-09 2009-10 2010-11 2011-12 2012-13 YTD Total
Companies 144 1496 279 1919
Individuals 66 55 392 1067 796 2376
Superannuation Funds 3 4 126 110 6 249
Total 69 59 662 2673 1087 4544

Source: ATO

4.48 The ATO attributes the increase in the default assessments it issued over the 2009-10 to 2011-12 financial year period to the expansion in scope and capability of its use of third party data to identify instances of unreported or under-reported income.67

4.49 In the 2012-13 year, the ATO planned to issue approximately 60,000 default assessment warning letters. Based on previous experience, it anticipated that 30,000 taxpayers would lodge after receipt of the letter and that 22,500 taxpayers may then be subject to default assessments. However, the YTD (as at 14 May 2013) data suggests there has been a decrease in the number of default assessments issued. This decrease may be due to a number of different factors, including the ATO's expanded use of data matching activities and associated compliance treatments as well as higher proportions of taxpayers responding to default assessment warning letters.

IGT conclusion — implemented

4.50 The IGT considers that the ATO has implemented this recommendation.

4.51 The IGT notes that with increased automation, the ATO's lodgement compliance work may become more efficient and address the risk of non-lodgement in a more timely manner.

4.52 The IGT also acknowledges that with greater use and expansion of the ATO's data matching program, opportunities may exist for the ATO to identify and engage with taxpayers through different compliance approaches other than default assessments.


49 IGT, Review into the Non-lodgement of Individual Income Tax Returns (2009) p 4.

50 ATO, 'Income Tax Lodgment — July 2009 — June 2010', internal ATO document, undated.

51 ATO, 'Income Tax Lodgement Approach and Treatment', internal ATO document, undated.

52 ATO, Lodgment Performance Income Tax Report (August 2012).

53 ATO, Individuals Segment Lodgment Performance Report (11th ed, May 2011).

54 DBM Consultants, Compliance with Income Tax Return Lodgment Obligations Research Final Report (June 2009); DBM Consultants, Compliance with Income Tax Return Lodgment Obligations Research Tax Agents and Businesses Quantitative Study (June 2010); DBM Consultants, Compliance with Income Tax Return Lodgment Obligations Research Individuals Quantitative Study (August 2010).

55 Colmar Brunton, Tracking the Individuals Segment (October 2009) p 75 and 76.

56 See for example: ATO, Community Perceptions Survey 2012 (June 2012) p 48; ATO, Micro Business Perceptions Survey 2012 (June 2012) p 8.

57 ATO, 'TPALS NRQL Analytical Model Evaluation', internal ATO document, September 2012.

58 ATO, communication to the IGT, 14 May 2013.

59 ibid.

60 ATO, 'Model Evaluation', above n 57.

61 ibid, p 23.

62 ATO, communications to the IGT, 12 October 2012 and 4 March 2013.

63 ATO, communication to the IGT, 12 October 2012.

64 Ibid, p 9.

65 ATO, communication to the IGT, 9 December 2013.

66 ATO, communication to the IGT, 14 May 2013.

67 ibid.