Show download pdf controls
  • Methodology

    We use a top-down approach to estimate the GST gap. This compares total GST liabilities for a financial year with the theoretical total GST liabilities for that year.

    The GST gap is the difference between the amount of GST payable under the law (theoretical GST liability) and the amount actually collected by us (actual GST revenue) for a defined period, typically a financial year.

    The gap can arise from a number of taxpayer behaviours. These include, but are not limited to:

    • non-reporting of GST
    • under-reporting of GST
    • over-claiming of GST credits.

    The gap estimate is made on the basis of the law as applicable for the relevant financial year. New or recent law changes will not be reflected in gap estimates.

    The methodology we have selected to estimate the net GST gap is outlined below, along with our assumptions, limitations, data sources and reliability rating as assessed by our independent expert panel.

    Find out about:

    Selecting the methodology

    Actual GST revenue is calculated on an accrual basis, matching revenue to the financial year in which the liability occurred, rather than to the financial year in which the revenue was received. The economic transaction method (ETM) is used, which is the revenue recognition basis applied for GST in the ATO financial statements.

    In selecting a methodology we researched and considered the full range of methodologies available, including top-down and bottom-up methods.

    We adopt a top-down methodology largely based on macroeconomic data from the ABS. This is considered the most suitable given the broad-based coverage of GST across goods and services and the high correlation between GST collections with consumer spending activity and other economic variables. We supplement the ABS data with a combination of ATO internal data and other government statistics.

    Applying the methodology

    The Australian GST system is a tax on final consumption expenditure on goods and services within the domestic economy. We apply the appropriate GST rate to the modified final consumption expenditure items which are taxable. This provides the theoretical tax liability, from which we subtract actual GST collections to obtain the tax gap estimates.

    We follow five steps in applying the top-down methodology to estimate the GST gap. The steps are summarised in Figure 4 and then described in detail.

    Figure 4: Steps to estimate the GST gap

    Figure 4: This image is a pictorial representation of the five calculation steps outlined in the text following this image. This image provides the names of the steps only.

    Find out about:

    Step 1: Construct theoretical GST base using expenditure data

    Starting with HFCE data subject to GST, we add estimates of expenditure for the following items:

    • new private dwellings investment expenditure (based on investment in new dwellings data)
    • consumers' share of ownership transfer costs
    • a proportion of land sales
    • net impact of international tourism.

    Step 2: Subtract GST concessions or exemptions

    We then remove specific expenditures included in HFCE. These are expenditures for which GST concessions or exemptions apply. These include:

    • expenditures that are exempt or concessionally taxed, such as food and education
    • input-taxed supplies, such as rent
    • certain financial supplies and reduced GST credits
    • concessions for entities with turnover less than $75,000 ($150,000 for not-for-profit entities).

    The residual amount is our estimate of total final consumption expenditure that is subject to GST.

    Step 3: Determine theoretical GST liability

    The total consumption expenditure value consists of the GST exclusive price and a GST equals to 10% of the price. We therefore isolate the total theoretical GST liability amount by dividing it by 11 (given the fixed GST rate of 10% incorporated in the data).

    Step 4: Calculate net gaps

    We subtract the actual GST liabilities reported on an accrual basis, including ATO compliance activities, from the theoretical total GST liability to estimate the net gap.

    Non-pursuable debt increases the net gap. Therefore, we add them back to the net gap estimate to come up with net gap with debt. This is the most accurate measure of tax gap for GST.

    Step 5: Calculate gross gaps

    The gross gap (including debt) is obtained by adding the liabilities raised from ATO compliance activities to the net gap figure.

    Some of the key assumptions of the methodology are as follows:

    • Final consumption expenditure represents all Australian consumption. No additional uplift for the black economy other than those incorporated into the ABS estimates has been applied. This is currently a 0.4% adjustment made to the HFCE estimate to account for under-reporting of sales.
    • There is no adjustment for inventory. Stock inventory levels are assumed to remain relatively constant.

    Summary of the estimation process

    Table 2 provides a summary of each step of the estimation process and the results for each year, from 2012–13 to 2017–18.

    Table 2: Summary of the GST gap estimation process

    Step

    Description

    2012–13

    2013–14

    2014–15

    2015–16

    2016–17

    2017–18

    1 to 3

    Total theoretical tax liability ($m)

    52,864

    55,899

    59,373

    62,667

    65,662

    68,555

    4

    Less final GST reported ($m)

    49,398

    52,808

    55,621

    58,169

    61,608

    64,218

    4.1

    Equals final GST liabilities not reported

    3,465

    3,091

    3,752

    4,498

    4,054

    4,337

    4.2

    Add non-pursuable debt ($m)

    593

    968

    487

    658

    658

    658

    5

    Net gap with debt estimate ($m)

    4,058

    4,059

    4,239

    5,156

    4,712

    4,995

    5.1

    Add compliance outcomes and taxpayer adjustments ($m)

    2,509

    2,497

    2,550

    2,279

    2,351

    2,282

    5.2

    Equals gross gap with debt estimate ($m)

    6,567

    6,556

    6,789

    7,435

    7,064

    7,276

    5.3

    Gross gap (%)

    12.4

    11.7

    11.4

    11.9

    10.8

    10.6

    5.4

    Net gap (%)

    7.7

    7.3

    7.1

    8.2

    7.2

    7.3

    Limitations

    The GST gap is estimated primarily using ABS Australian National Accounts data. The reliability of the gap estimate therefore depends on the accuracy and completeness of that data. National Accounts data includes a margin of error and imposes some limitations on the estimate. Specific issues include:

    • Sampling and non-sampling errors may exist.
    • Underlying data is subject to revision, which can vary historical trend results and the estimated GST gap.
    • Timing differences can exist between the National Accounts and GST treatment for certain supplies.

    There is a time lag between the completion of a period and the publishing of Australian National Accounts data.

    In addition, concessions and exemptions are identified and estimated in the Treasury Tax Expenditures Statement. The statement estimates can have a wide range and are not exhaustive, with only major exemptions and exceptions identified.

    Accounting for black economy activity in a top-down model

    Due to the GST gap being calculated using the National Accounts, accounting for black economy activity can only be estimated at a macro level. What this means, is that we look to calculate the tax impact of the black economy based on the ABS stated level of 1.5%. Therefore, the black economy impact is 1.5% of the total theoretical liability. This translates into a tax impact of approximately $1.01 billion. This estimate is contained within the total theoretical liability calculation.

    Updates and revisions to previous estimates

    Periodically, information collated and analysed by the ABS is updated or revised. For the GST gap estimate, this means that changes to the underlying data will have an effect on the outcomes of our analysis. The ABS generally revise their annual benchmarks for the prior three years, however they periodically revise the entire time series. The effect of these changes upon our estimate is demonstrated at Figure 5.

    Figure 5: Effect of ABS and National Accounts revisions on previous GST gap estimates, 2011–12 to 2018–19

    Figure 5: This graph provides a visual representation of the previous and current net gap estimates provided at Table 3.

    Table 3 shows that the gap percentages are similar for the years we have measured.

    Table 3: Summary of published GST net gap percentages

    Gap release year

    2011–12
    (%)

    2012–13
    (%)

    2013–14
    (%)

    2014–15
    (%)

    2015–16
    (%)

    2016–17
    (%)

    2017–18
    (%)

    2015

    7.7

    7.1

    6.5

    na

    na

    na

    na

    2016

    7.0

    5.8

    6.1

    6.5

    na

    na

    na

    2017

    7.0

    5.7

    6.1

    6.7

    7.3

    na

    na

    2018

    8.1

    7.4

    7.1

    7.4

    8.7

    7.9

    na

    2019

    8.3

    7.7

    7.3

    7.1

    8.2

    7.2

    7.3

    The GST gap estimate may not reflect the true level of non-compliance in the GST system, as there are likely to be taxpayers who over-comply. This may occur where taxpayers under-claim GST credits they are entitled to, or incorrectly classify supplies as subject to GST when they are actually GST-free. As a result, the real revenue gap using a top-down methodology may be understated.

    We have not produced a confidence interval for the estimate of the GST gap. Each of the assumptions used in the top-down model has its own errors, many of which are unquantifiable.

    Data sources

    We adopt a top-down methodology to estimate the GST gap using a combination of ATO and external data, such as:

    • ABS Australian National Accounts
    • ABS Household Expenditure Survey
    • Treasury Tax Expenditures Statement.

    See also:

    Reliability

    The ATO estimate of the GST gap has been assessed by an independent expert panel. The process is described in Principles and approaches to measuring gaps.

    Based on advice from the independent expert panel, the reliability rating for this estimate is medium.

    Data used to estimate the theoretical tax liability is largely based on external data sources and incorporates a factor for the black economy.

    The methodology is based on the approach recommended by the IMF and applied to value added tax (VAT) / GST gap estimation globally. We retain some concerns over the completeness of the tax base covered, and the estimates for concessional and exempt expenditure.

    Figure 6: Reliability rating scale from very low to very high – GST gap

    Figure 6: This image is a graphical representation of the reliability rating for the current large super funds income tax gap estimate. It graphically represents a rating of medium, which is a score between 16 and 20. The maximum score is 30.

    Return to:

    See also:

      Last modified: 17 Oct 2019QC 57175