Show download pdf controls
  • Gap estimation

    The tax gap refers to the difference between what the ATO collects and the amount that would have been collected if every taxpayer was fully compliant.


    For our purposes, we define the gap as the difference between the estimated amount of a liability or obligation theoretically payable and the amount actually reported to, or collected by, revenue authorities over a defined period. The gap includes amounts that are incorrectly reported, as well as amounts that we do not expect will ever be paid.

    Our estimates solely reflect the gap with respect to compliance with the law and administrative approaches at the time. The gap includes tax evasion, fraud, incorrect reporting, non-payment of liabilities, non-registration and non-lodgment. It does not include interest charges or penalties.

    In estimating gaps, we look at four key elements:

    • the population (the individuals, groups and organisations) that should pay
    • the total amount the population should pay (theoretical total liability)
    • the total amount the population is actually recorded as being liable to pay
    • the amounts we do not expect will ever be paid – recorded liabilities determined to be irrecoverable at law or not economical to pursue (for example, insolvency debts).

    Our estimates do not include tax forgone as a result of policy decisions, such as exempting certain types of income or providing concessions to certain groups of taxpayers. These are sometimes described as 'tax expenditures' and are the subject of separate research and analysis undertaken annually by Treasury. The estimates also exclude tax minimisation which is allowable under the law.

    We have estimates of both the dollar and percentage value of:

    • the gross tax gap – the gap prior to our compliance activities, and
    • the net tax gap – the gap after our compliance activities.

    Dollar gap estimates are converted to percentages by dividing the net or gross gap by the theoretical total liability.

    Methodological approaches

    Our tax gap estimates are derived from a wide range of sources, including publicly-available information and ATO administrative data. Broadly, we apply either a top-down or bottom-up approach to estimating each gap.

    Top-down approaches use externally-provided aggregated data sources to estimate the size of the tax base, from which we estimate theoretical tax liability. The difference between the theoretical tax liability and the amount we receive is the estimated tax gap. A top-down approach is typically used for indirect taxes.

    Bottom-up approaches involve a detailed examination of data sources, such as tax returns, audit results, risk registers or third-party data-matching information. We then extrapolate the results to determine the extent of non-compliance across the whole population, from which we estimate the tax gap. This approach generally involves applying statistical techniques to determine the incidence and value of non-compliance. A bottom-up approach is typically used for direct taxes.

    Ideally, a combined approach is used, as each methodology complements the other and improves the accuracy and utility of the estimates.

    The table below shows the type of methodological approach we currently use for each tax gap estimate.

    Methodological approach used, by tax gap

    Gap estimate

    Methodological approach

    Goods and services tax (GST)

    Top down

    Luxury car tax (LCT) payable

    Bottom up (data matching)

    Wine equalisation tax (WET) payable

    Top down

    Petrol and diesel excise and duty

    Top down

    Pay as you go (PAYG) withholding

    Top down

    Fuel tax credits (FTC)

    Bottom up (random enquiry program)

    We choose the methodology that provides the most reliable estimate. In general, data availability and quality are the key deciding factors for the approach used. We use ATO data and a wide variety of publicly available information and proprietary data to estimate gaps. In general, our estimates cover the 2009–10 to 2014–15 income years. In some cases, data limitations and lags mean that full coverage is not possible. Projections are made where data is limited.

    We assess our methodologies for reliability to select the most appropriate methodology and to test our results.

    Reliability assessment

    All gap estimates are assessed for reliability against ten criteria. The reliability rating provides a transparent assessment of our gap estimates, drawing on International Monetary Fund and ATO expertise. We summarise this in a rating assessment for each gap estimate.

    The ten criteria are considered of equal importance. They are:

    Estimation framework

    To what extent does the estimate:

    • capture the appropriate tax base
    • cover all potential taxpayers
    • account for all potential forms of non-compliance
    • avoid overlap between any two components of the framework.


    To what extent does the estimate:

    • meet International Monetary Fund methodology criteria
    • use multiple approaches that are validated internally and against accepted international standards
    • sensitivity test for underlying changes in the model, structure and assumptions
    • evaluate and assess assumptions, judgment or expertise used.

    Internal processes and delivery

    To what degree is the estimate evaluated for:

    • reliability and repeatability of data and documentation
    • testing, evaluation and measurement against other sources, both internal and external.

    For each estimate, each criterion is scored from 0 ('does not meet in any way') to 3 ('totally meets'). The unweighted sum of these scores determines the reliability rating. The total reliability score ranges from 0 to 30 and is placed into one of four categories, as follows:

    • Very low (a score of 0 to 5) – The estimate results are preliminary or interim results, often being pilot estimates in their first year of production. The estimate may have a number of issues, such as high margins of error, which affect the integrity of the estimate. The estimate is not confirmed by other independent analysis.
    • Low (a score of 6 to 15) – The estimate has a broad number of factors that are not considered. The estimate has material margins of error. Estimates may be partially confirmed by other analyses. Improvements when made may significantly alter the gap estimates.
    • Medium (a score of 16 to 25) – A small number of factors are not considered. If addressed, these factors may change gap estimates to a limited or immaterial degree. The estimate has acceptably low margins of error. Estimates are potentially derived from more than one calculation. It is materially confirmed by other analyses such as risk models and intelligence scans.
    • High (a score of 26 to 30) – All factors are considered. The estimate has very low margins of error. Possibly derived from more than one calculation, the estimate is confirmed by other analyses, such as risk models and intelligence scans. The estimates can provide detailed insights into the levels of compliance across the population.

    The reliability assessment of each estimate, as endorsed by our expert panel, is shown in the following table.

    Reliability assessment of tax gap estimates

    Gap estimate


    Goods and services tax


    Luxury car tax


    Wine equalisation tax (WET)


    Petrol and diesel excise and duty


    PAYG withholding


    Fuel tax credits


      Last modified: 21 Sep 2020QC 50394