We use a 4-step top-down methodology to estimate the GST gap.
Step 1: Construct theoretical GST base using expenditure data
Starting with the Australian Bureau of Statistics (ABS) estimate of household final consumption expenditure (HFCE), we add estimates of expenditure for the following items which are subject to the GST:
- new private dwellings investment expenditure (based on investment in new dwellings, alterations and additions)
- consumers' share of ownership transfer costs
- a proportion of land sales.
Step 2: Subtract spending where GST concessions or exemptions apply
We then remove specific expenditures included in the ABS measure of HFCE but for which GST concessions or exemptions apply. These expenditures are not part of the theoretical GST base.
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 the theoretical GST base subject to GST.
Step 3: Determine theoretical GST liability
The total theoretical GST base estimated above consists of the GST exclusive price and the appropriate GST (which equals 10% of the GST exclusive price).
We estimate the total theoretical GST liability by dividing the theoretical GST base by 11 (given the fixed GST rate of 10% is incorporated in the estimated GST base).
Step 4: Consolidate the gap estimates
We subtract the actual GST liabilities reported on an accrual basis, including our compliance activities, from the theoretical total GST liability to estimate the net gap.
Non-pursuable debt is considered part of the net gap. Therefore, we add this amount back to estimate the net gap including debt. This is the most accurate measure of the tax gap for GST.
We obtain the gross gap (including debt) by adding the liabilities raised from our compliance activities to the net gap estimate.
Some of the key assumptions of the methodology include:
- HFCE represents all Australian consumption by households. No additional uplift for the shadow economy has been applied other than those incorporated into the ABS estimates. The ABS currently makes an upwards adjustment to a number of the HFCE components to account for under-reporting of sales. These adjustments made by the ABS aggregate to around 0.4% of their HFCE estimate.
- No adjustments have been made for timing issues in some of the National Accounts aggregates used to quantify total theoretical GST revenue, despite some known conceptual misalignments in private dwelling investment.
- The estimates of each HFCE component not subject to GST are derived from various Tax Benchmarks and Variations Statements and Household Expenditure Surveys. The compositional shifts in spending caused by the COVID-19 pandemic are quite dynamic and may not be adequately reflected in the data.
Summary of the estimation process
Table 2 provides a summary of each step of the estimation process and the results for each year, from 2015–16 to 2020–21.
Step | Description | 2016–17 | 2017–18 | 2018–19 | 2019–20 | 2020–21 | 2021–22 |
---|---|---|---|---|---|---|---|
1 to 3 | Total theoretical tax liability ($m) | 64,861 | 67,879 | 69,879 | 67,554 | 70,622 | 78,264 |
4 | Less final GST reported ($m) | 61,811 | 64,308 | 65,630 | 65,798 | 70,041 | 77,750 |
4.1 | Equals final GST liabilities not reported | 3,050 | 3,570 | 4,249 | 1,756 | 581 | 514 |
4.2 | Add non-pursuable debt ($m) | 672 | 695 | 768 | 1,274 | 1,359 | 2,331 |
4.3 | Net gap with debt estimate ($m) | 3,722 | 4,265 | 5,017 | 3,031 | 1,939 | 2,845 |
4.4 | Add compliance outcomes and taxpayer adjustments ($m) | 2,599 | 2,672 | 2,869 | 2,175 | 3,044 | 5,909 |
4.5 | Equals gross gap with debt estimate ($m) | 6,321 | 6,937 | 7,887 | 5,206 | 4,983 | 8,754 |
4.6 | Gross gap (%) | 9.7 | 10.2 | 11.3 | 7.7 | 7.1 | 11.2 |
4.7 | Net gap (%) | 5.7 | 6.3 | 7.2 | 4.5 | 2.7 | 3.6 |
Find out more about our overall research methodology, data sources and analysis for creating our tax gap estimates.
Limitations
The GST top-down tax gap model relies on various ABS economic aggregates to construct a proxy for the GST base. The reliability of the gap estimates therefore depends on the accuracy and completeness of that data. National Accounts data includes a margin of error and imposes some limitations on gap estimates.
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.
The latest ABS national accounts data were released in October 2022. The ABS made a number of revisions to their previous estimates of HFCE. Some components were revised higher and others lower; but in aggregate HFCE was increased. These revisions therefore overall increased the:
- theoretical GST base
- theoretical GST liability
- estimated tax gaps.
This impact alone increased the estimated tax gaps for 2020–21 by around $320 million.
In addition, concessions and exemptions are identified and estimated in the Treasury Tax Benchmarks and Variations Statement. The statement estimates can have a wide range and are not exhaustive, with only major exemptions and exceptions identified.
Accounting for shadow economy activity in a top-down model
The theoretical GST liability is based on macro aggregates compiled by the ABS. Where relevant, these have been adjusted upwards to include an indicative impact of the shadow economy.
Reflecting this, we make no further adjustment for the shadow economy when calculating the theoretical GST liability and resulting tax gap. This means our estimates include a component due to the shadow economy.
Estimating the size of the shadow economy can only be done with a significant margin of error. In times of economic shock such as we have experienced during COVID-19 the range of uncertainty relating to such estimates is likely to increase. The current approach by the ABS assumes that the non-observed economy is around 0.4% of HFCE; with that markup relative to HFCE constant over time. To the extent that this assumption does not hold, this can have a material impact on estimates of consumption and in turn our GST gap estimates.
Updates and revisions to previous estimates
Each year we refresh our estimates in line with the annual report. Changes from previously published estimates occur for a variety of reasons, including:
- improvements in methodology
- revisions to data
- additional information becoming available.
Periodically, information collated and analysed by the ABS is updated or revised. For the GST gap estimate, this means changes to the underlying data will influence the outcomes of our analysis.
The ABS generally revise their annual benchmarks for the prior 3 years. However, they periodically undertake more comprehensive reviews resulting in more extensive revisions.
The effects of these changes to methodology and underlying data on our gap estimates are demonstrated at Figure 2.
Figure 2: Effect of ABS revisions and methodological changes on previous GST net gap estimates, 2010–11 to 2021–22
This data is set out in Table 3, shown as a percentage of theoretical GST revenue.
| 2010–11 | 2011–12 | 2012–13 | 2013–14 | 2014–15 | 2015–16 | 2016–17 | 2017–18 | 2018–19 | 2019–20 | 2020–21 | 2021–22 |
---|---|---|---|---|---|---|---|---|---|---|---|---|
2023 Program | n/a | n/a | n/a | n/a | n/a | 8.3 | 5.7 | 6.3 | 7.2 | 4.5 | 2.7 | 3.6 |
2022 Program | n/a | n/a | n/a | n/a | n/a | 8.8 | 6.3 | 6.8 | 7.7 | 6.3 | 5.9 | n/a |
2021 Program | n/a | n/a | n/a | 8.4 | 8.9 | 10.2 | 8.4 | 7.8 | 7.8 | 7.8 | n/a | n/a |
2020 Program | n/a | n/a | n/a | 7.3 | 7.5 | 8.2 | 6.9 | 7.3 | 8.1 | n/a | n/a | n/a |
2019 Program | n/a | n/a | 7.7 | 7.3 | 7.1 | 8.2 | 7.2 | 7.3 | n/a | n/a | n/a | n/a |
2018 Program | n/a | 8.1 | 7.4 | 7.1 | 7.4 | 8.7 | 7.9 | n/a | n/a | n/a | n/a | n/a |
2017 Program | 7.1 | 7.0 | 5.7 | 6.1 | 6.7 | 7.3 | n/a | n/a | n/a | n/a | n/a | n/a |
2016 Program | 7.0 | 7.0 | 5.8 | 6.1 | 6.5 | n/a | n/a | n/a | n/a | n/a | n/a | n/a |
2015 Program | 6.8 | 7.7 | 7.1 | 6.5 | n/a | n/a | n/a | n/a | n/a | n/a | n/a | n/a |