Show download pdf controls
  • Methodology

    To be eligible for a fuel tax credit, a taxpayer must be registered for goods and services tax (GST) at the time it acquires, manufactures or imports taxable fuel for use in carrying on a business.

    The fuel tax credit gap is the difference between the estimated value of fuel tax credits claimants are entitled to claim according to the law and the value of fuel tax credits they actually claimed in a year.

    Here we outline the methodology we have selected to estimate the fuel tax credits gap. We detail our assumptions, limitations, data sources and reliability rating as assessed by our independent expert panel.

    Find out about:

    See also:

    Selecting the methodology

    Following consultation with our independent expert panel, we selected a bottom-up model based methodology.

    We used data derived from our random enquiry program (REP) undertaken for the 2014–15 financial year. This program randomly selected taxpayers who claimed fuel tax credits on their business activity statements. We reviewed their reporting to detect issues and identify non-compliance. We then split the sample into two strata of fuel tax credit claims, above and below $10,000 for the year, and by broad industry groups.

    Under the REP, we assume that:

    • the sample is representative of the population
    • the incidence and magnitude of non-compliance found in the random sample is representative of the population
    • the final changes to liabilities as a result of compliance activities are correct at law and not subject to further change.

    Applying the methodology

    This approach first determines the value of non-compliance for the sample. We then apply the mean amendments and incidence rate to the fuel tax credits population for 2014–15. From this, we calculate a ratio which we apply to other years to estimate the gap. This is the modified approach to extrapolation. It is a change to the method we used in the 2018 release year.

    The seven steps we used to calculate the gap are summarised in Figure 3 and then described in detail.

    Figure 3: Steps to estimate the fuel tax credits gap

    Figure 3: This image is a pictorial representation of the seven calculation steps outlined in the text following this image. This image provides the names of the steps only.

    Find out about:

    Step 1: Estimate unreported amounts and extrapolate to population

    Calculate the mean amendment and incidence rates for underclaims and overclaims found in the REP and extrapolate to the 2014–15 population.

    Step 2: Calculate ratio of extrapolation to total amendments for 2014–15 and apply to each year

    Take the extrapolated amount from Step 1 and divide by the total amendment value to obtain a ratio.

    Take the ratio and multiply by the total amendment value for the respective year.

    Step 3: Estimate for non-detection

    Multiply the amount from Step 2 by the uplift factor to account for non-detection.

    Step 4: Estimate for non-pursuable debt

    Add non-pursuable debt to account for over claims not repaid.

    Step 5: Estimate the gross gap

    Add Steps 2 to 4 to estimate the gross gap.

    Step 6: Estimate the net gap

    Subtract amendments from the gross gap to estimate the net gap.

    Step 7: Estimate the theoretical credit

    Subtract the gross gap from voluntary credits claimed to estimate the theoretical credit.

    Summary of the estimation process

    Table 2 displays the dollar values at each step rounded to the nearest million.

    Table 2: Summary of estimation process for steps of the fuel tax credits gap

    Step

    Description

    2014–15

    2015–16

    2016–17

    2017–18

    1

    Estimate unreported amounts and extrapolate to 2014–15 population (A) ($m)

    −29

    na

    na

    na

    2

    Divide (A) by total value of amendments (B) for 2014–15 to obtain a ratio (C) ($m)

    1.09

    1.09

    1.09

    1.09

    3

    For each year multiply total value of amendments (B) by the same ratio (C) ($m)

    −29

    −48

    −41

    −40

    4

    Add estimate for non-detection (D) ($m)

    −7

    −12

    −10

    −10

    5.1

    Add claims made incorrectly and not paid back (E) ($m)

    8

    8

    8

    8

    5.2

    Equals gross gap (F) ($m)

    −29

    −53

    −44

    −42

    6.1

    Minus amendments (B) ($m)

    −27

    −44

    −38

    −36

    6.2

    Equals net gap (H) ($m)

    −2

    −9

    −6

    −6

    7.1

    Estimate theoretical credit by subtracting the gross gap (F) from total FTC claims ($m)

    6,081

    6,155

    6,335

    6,935

    7.2

    Gross gap (%)

    −0.5%

    −0.9%

    −0.7%

    −0.6%

    7.3

    Net gap (%)

    0.0%

    −0.1%

    −0.1%

    −0.1%

    Note: Claims made incorrectly and not paid back are amounts that are written off as irrecoverable at law or not economical to pursue.

    Return to:

    Limitations

    The gap estimates are derived from the randomly selected sample. Although it is assumed that the sample is representative of the population, there may be isolated instances of non-compliance around fuel tax credits outside the sample, which have not been captured in this analysis. It is also assumed that the ratio of the extrapolation to amendments is relatively constant over time.

    The sample size for this random enquiry program (REP) is small. In general terms, the larger the sample the more reliable the results.

    The extent of non-detection is unknown and is extremely challenging to measure. The REP is subject to uncertainty around the impact of the non-detection error.

    This estimate does not include the population that may be entitled to fuel tax credits, but have not registered and lodged a claim to receive fuel tax credits.

    This population does not include a small number of taxpayers who claimed for fuel tax credits for domestic electricity generation.

    Return to:

    Updates and revisions to previous estimates

    Figure 4 displays the net gap from our current estimate, which uses the modified approach, compared to our previous estimate published in 2018.

    Figure 4: Current and previous net gap estimates, 2011–12 to 2017–18

    Figure 4: This graph provides a visual representation of the previous and current net gap estimates provided at Table 3.

    For the estimate released in 2018, the change in 2014–15 from using the benchmark to the REP results led to the gap moving from positive to negative in that year. The current estimate, using the modified approach for each year from 2014–15 to 2017–18, produces a more stable series. These features are summarised in Table 3.

    Table 3: Summary of published fuel tax credits net gap percentages

    Gap release year

    2011–12
    (%)

    2012–13
    (%)

    2013–14
    (%)

    2014–15
    (%)

    2015–16
    (%)

    2016–17
    (%)

    2017–18
    (%)

    2018

    0.7

    0.7

    0.8

    −0.4

    −0.2

    −0.3

    na

    2019

    na

    na

    Na

    0.0

    −0.1

    −0.1

    −0.1

    Note: The estimate released in 2018 does not cover 2017–18. Similarly, the current estimate does not cover the years 2011–12 to 2013–14.

    Data sources

    Five ATO data sources are used to calculate the fuel tax credit tax gap estimate:

    1. ATO fuel tax credits taxation population statistics as published on data.gov.auExternal Link (both numbers of taxpayers and value of claims including Industry statistics). See Australian Taxation Office Taxation Statistics 2016–17 Excise and Fuel Schemes Table 1 External Linkand Table 4External Link.
    2. 2014–15 to 2017–18 Commissioner of Taxation annual reports into administered programs of both tax reported and net gaps.
    3. ATO compliance data. Final results of fuel tax credit compliance cases of taxpayers claiming fuel tax credit for the 2010–11 to 2016–17 financial years.
    4. 2014–15 REP results, which are derived from the compliance checks on the random sample of taxpayers who claimed fuel tax credits.
    5. ATO benchmarking study conducted on 2006–07 claims. This was used to calculate the estimates for 2011–12 to 2013–14.

    The ATO benchmarking study was undertaken in 2007–08 and provided diesel fuel benchmarking data that was used for estimating the gaps up to 2013–14.

    The 2014–15 REP reviewed the 2014–15 claims of 271 randomly selected fuel tax credit taxpayers. The data from this second program has been used to estimate the 2014–15 onwards fuel tax credits gaps.

    The program initially selected 500 claimants, split into two-tiered strata: claims above and below $10,000, and by broad industry group. Each taxpayer was reviewed and 271 were selected, based on risk, for a desk-based audit or review.

    The largest taxpayers for fuel tax credits were excluded from the scope of the REP. We have assumed our relationship management program provides a level of assurance over the integrity of accounting systems in place to cover this population.

    Reliability

    Our estimate of the fuel tax credits gap has been assessed by an independent expert panel, as described in Principles and approaches to measuring gaps.

    Based on advice from the independent expert panel, the reliability rating for the fuel tax credits gap estimate now sits at a lower level within the medium range. This is as a result of changing to the modified approach, which required an additional assumption.

    We take the results from the random enquiry program (REP) and project those results over the total lodged population of taxpayers, including those who lodge late. We take the ratio of this extrapolation to total amendments and apply it to each year. This method looks at all items on the activity statement, and the taxpayer information we have received. Non-payment is also addressed.

    The key assumption with the REP is that the observations of the sample apply to the population. We have stratified the sampling process to ensure it is representative. We have assumed that the ratio of the extrapolation to total amendments is relatively constant over time.

    Future work that can increase reliability includes:

    • exploring a top-down estimate which could be used for benchmarking
    • investigating trends in fuel usage to help better understand the gap.

    Figure 5: Reliability rating scale from very low to very high – fuel tax credits gap

    Figure 5: This image is a graphical representation of the reliability rating for the current Fuel excise gap estimate. It graphically represents a rating of medium, which is a score between 16 and 20. The maximum score is 30.

    Return to:

      Last modified: 17 Oct 2019QC 57176