5.1 The second term of reference involves an examination of the current time frames which apply to the processing of various types of GST refunds and of the performance standards which the Tax Office uses in relation to issuing GST refunds on a timely basis. This chapter sets out the Inspector-General's findings and recommendations in relation to this term of reference.

5.2 A summary of Tax Office processes relevant to GST refunds is set out in Appendix 5.

Current Tax Office time frames for processing of routine refunds

5.3 During the course of this review a major concern raised by taxpayers and tax practitioners was that the Tax Office took too long to issue GST refunds generally and, in particular, took too long to issue a refund where the refund was subject to certain verification checks.

5.4 The Commissioner of Taxation has stated publicly that routine electronically lodged GST refund claims (that is, refunds that are not stopped for verification or other reasons) should be paid within four days of lodgment. In a speech given on 15 April 2004, the Commissioner stated that, with new enhancements that were to be made to the Business Portal from mid-2004, taxpayers who lodged electronically would be able to receive real time confirmation of the processing of their statement and that their refund would then be paid within three working days of that notification.35

5.5 The Tax Office has also stated to the Inspector-General that routine refunds claimed via a paper return will be paid within six working days of lodgment.36

5.6 For routine GST refunds, the Tax Office was only able to provide the Inspector-General with performance statistics for the time taken for all GST refunds to reach the stage of having been assessed by the RRE. The Tax Office does not prepare performance statistics for the time taken after clearance for payment by the RRE or by a verification area of the Tax Office. This part of the refund process occurs within the Operations business line of the Tax Office and is known as the 'refunder process'. However, according to the Tax Office this process should take one day to complete for routine GST refunds. A further two-day period is then required by the banking system to deposit the refund into the taxpayer's bank account.

5.7 Based on the advice that the refunder process will usually take three days to complete, the material that was provided to the Inspector-General shows that, during 2003-04, most (that is, 98 per cent of) electronically lodged routine BAS refund claims were paid to taxpayers within six working days of lodgment of the claim.37 On a calendar day basis, this time period equates to a minimum period of eight days (including a weekend). During that year, the overwhelming majority (that is, 91 per cent) of routine BAS refund claims lodged by paper returns were paid within 10 working days of lodgment.38 This 10 working day period equates to 14 calendar days.

5.8 The Inspector-General notes that the Tax Office's 2003-04 annual report indicates that it is endeavouring to improve current processing times for refunds, and to therefore improve the service it provides to taxpayers. One future initiative in this regard will be the increased automation of its refund process.39

5.9 The Inspector-General also notes that the Tax Office currently publishes certain weekly activity statement processing statistics on its website. However, these only indicate the percentages of activity statement refunds that are processed within 14 days. They do not provide a break-up between GST refunds and other activity statement refunds claimed, or between paper refund returns and electronic refund returns.

5.10 This leads to the following recommendation.

Subsidiary recommendation 2

The Inspector-General recommends that the Tax Office publish statistics to advise taxpayers, on a regular basis, of the number of days it will take the Tax Office to pay a GST refund after lodgment of either a paper or electronic BAS.

Tax Office response

5.11 The Tax Office agrees with this recommendation. The Inspector-General acknowledges that the Tax Office currently publishes weekly activity statement processing statistics on the percentages of activity statement refunds that are processed within 14 days on our website. In addition to these statistics, we will publish the average expected time to pay a routine activity statement refund. The information categorised below based on prior week's turnaround time will be published on a weekly basis.

  • Paper lodgments
    • Original and revision business activity statements
    • Original and revision instalment activity statements
  • Electronic lodgments
    • Original and revision business activity statements
    • Original and revision instalment activity statements

Current time frames for processing of non-routine refunds

5.12 According to material that has been published by the Tax Office, there are five main reasons why GST refunds may be delayed. Refunds which have been delayed for one of these five reasons are referred to as non-routine refunds in this report.40

5.13 These reasons are as follows:

  • another activity statement for the taxpayer is overdue or is still being processed;
  • the taxpayer has not provided to the Tax Office details of the bank account into which the refund can be paid;
  • a previous electronic funds transfer refund has been rejected for the taxpayer;
  • the taxpayer's account has an intercept in progress for either the refund in question, another refund or another assessment; or
  • the Tax Office has decided to confirm the accuracy of the GST refund prior to its payment.

5.14 This list does not include a sixth possible reason — the operation of edit, scanning and exception tests which are conducted within the processing area of the Tax Office prior to a paper-lodged refund being processed electronically. This review has ascertained that these checks may result in a refund being delayed, sometime for several days.

5.15 The first four reasons for a GST refund being delayed and the additional sixth reason involve activities conducted by the Operations business line of the Tax Office. This is the business line responsible for the processing of GST refunds.

5.16 The fifth reason why a GST refund may be delayed — that it is being held up for verification — involves activities which are carried out by one of the compliance/auditing areas of the Tax Office.

5.17 If the proposals referred to earlier in this report are fully implemented by the Tax Office, GST refunds for many large enterprise taxpayers and government taxpayers may no longer be delayed due to pre-issue verification checks.

5.18 Delays in GST refund processing caused by pre-issue verification checks may also no longer arise for many small business taxpayers.

5.19 The following material discusses the reasons for possible GST refund delays which may continue to apply to certain GST taxpayers in the future under the following two broad headings:

  • processing activities which may continue to cause a delayed GST refund; and
  • verification activities which may continue to cause a delayed GST refund.

Processing activities which may delay a GST refund

Overdue and revised activity statements

5.20 A GST refund will not be paid where another activity statement is outstanding, or is lodged but not yet processed. The Tax Office has advised that once all overdue activity statements are processed the refund will automatically issue, provided that all other conditions are met.

5.21 Tax Office procedure is to pursue any outstanding activity statements, initially by letter, escalating to telephone contact where lodgment of the outstanding statements is not forthcoming.

5.22 The Tax Office has advised that, at any single point in time, approximately one-third of the GST refund cases that are held up because they are awaiting processing within the 'refunder' area of the Tax Office are delayed because of outstanding activity statement lodgments.41

5.23 In examining cases during the course of this review, the Inspector-General found that GST refunds were being delayed under this heading in situations where the outstanding activity statement was an activity statement which revised prior period amounts.

Tax Office response made during the review

5.24 The Tax Office has recognised that there are no lodgment issues associated with revised business activity statements and therefore no need to hold refunds in these circumstances. As a result, in September 2004 changes were made to the refund process to exclude revision activity statements from consideration when determining whether a refund should be released.42

Lodgment processes for revised BASs

5.25 A significant percentage of delayed refund cases that were sampled by the Inspector-General during the course of this review involved revised credit activity statements.

5.26 Up until August 2004, revisions to activity statements could not be made by electronic means, even where the refund was large and involved a large company or other enterprise. This meant that revision statements for all taxpayers needed to be in either paper form or by way of a letter.

5.27 Paper revision forms are subject to the same edit, scanning and exception tests that apply to all paper-lodged activity statements. Where revisions are lodged by letter, the figures in the revision letter need to be manually keyed into the Tax Office's computer systems. All these processes create the potential for delays.

Tax Office response made during the review

5.28 During the course of this review, the Tax Office has advised that it has now introduced a process under which a revised BAS can be lodged electronically. It advises that this should have the result of reducing the volumes of all revisions, including credit revisions, that are sent on paper and which may require manual keying and other forms of manual intervention.

5.29 The Tax Office has also advised that new procedures have been introduced so that credit revisions that are set out on paper are identified, separated out from other correspondence and given priority actioning.43

Absence of bank account details or incorrect bank account details

5.30 The Tax Office has advised that generally refunds are not able to be paid if bank account details are not recorded. Refunds are generally required under the law to be paid directly into a taxpayer's bank account. There are, however, a limited number of situations where these details are not required because the Tax Office has agreed to pay the GST refund by way of cheque.44

5.31 Where bank account details are required for payment of the refund, a letter to the taxpayer is automatically generated requesting the provision of these details.45 Once the financial institution details are updated, the Tax Office has advised that the refund will automatically issue, provided that all other conditions are met.

5.32 The Tax Office has advised that, as at June 2004, approximately two-thirds of the refunds that were held up during the refunder process were held up due to the absence of bank account details.46

Taxpayer concerns

5.33 In submissions made to this review, taxpayers raised a number of concerns about the Tax Office's failure to pay a GST refund unless bank account details for the particular taxpayer had been provided.

5.34 Firstly, taxpayers submitted that there was no procedure whereby a taxpayer who lodged a GST refund for the first time could notify the Tax Office of their bank account details on the BAS form itself. These taxpayers submitted that the BAS form should be amended so as to include a space which allowed for them to provide or update their bank account details.

5.35 Secondly, one submission noted an instance where the Tax Office had paid a GST refund into a wrong bank account. As a result, the taxpayer originally entitled to the refund had experienced a significant delay in receiving their correct refund entitlement.

5.36 Thirdly, submissions noted that the Tax Office had adopted a policy of requiring that the bank account into which any refund was paid must be in the name of the taxpayer who was entitled to the refund. This had caused a number of refunds, such as those claimed by subsidiaries in large company groups or by funds being managed by a single fund manager, to be delayed. This delay arose because the large company group or fund manager operated only one bank account for all of its subsidiaries or funds under management.

Tax Office response made during the review

5.37 The Tax Office has advised that, from August 2004, taxpayers are able to update their bank account details on-line via the business and tax agent portals. The Inspector-General notes that this initiative will go some way towards addressing one of the above concerns raised by taxpayers.

5.38 The Tax Office has also advised that it has concerns about the ability for bank account details to be changed via the BAS. It has stated that the ability to change bank account details by someone completing the form presents an unacceptable fraud risk, especially for large business entities.47

Inspector-General's observations

5.39 The Inspector-General notes that these comments by the Tax Office may be based, in part, on a lack of awareness of the nature of controls within large business entities which apply to BAS forms which may mitigate this Tax Office concern.

5.40 The Inspector-General also notes that the concerns raised about the need to provide bank account details for the particular entity to which the refund is to be paid appear to have been rectified by the issue of Practice Statement PS LA 2004/7. This practice statement was issued early in the course of this review.

5.41 The Inspector-General notes that the Tax Office has provided detailed instructions to its staff on how to deal with reclaiming refunds paid to incorrect accounts. These guidelines appear to adequately address the issue of ensuring that a quick remedy is provided to taxpayers whose refunds have not been paid because they have been paid into another taxpayer's account.48

Existence of intercepts

5.42 The existence of intercepts (being situations where a refund is stopped because of an indicator or notation on a taxpayer's account with the Tax Office) is a further reason for a GST refund being delayed.

5.43 Intercepts which result in the offset of the GST refund against other outstanding debts are the subject of the fourth term of reference for this review and are discussed in the next chapter.

5.44 One particular circumstance involving an intercept other than an offset was raised in submissions made by taxpayers to this review. This was where the automatic processing of a refund was stopped because the taxpayer had lodged a request for amendment to one or more previous income tax assessments. The Inspector-General was advised that a GST refund could be stopped in this case even though the relevant amendment sought to decrease, not increase, the amount of income tax previously paid.

5.45 This leads to the following subsidiary recommendation.

Subsidiary recommendation 3

The Inspector-General recommends that the Tax Office take steps to ensure that credit amendments to income tax assessments do not inappropriately halt the payment of GST refunds.

Tax Office response

5.46 The Tax Office agrees with this recommendation. However, our current processing systems do not enable differentiation between credit and debit amendments, therefore it is not possible to release a GST refund without waiting for an income tax amendment to be finalised and the transactions posted to the account.

5.47 We acknowledge this deficiency in our current system and will take steps to ensure that these issues are addressed with the introduction of our new processing systems in January 2008.

5.48 As an interim arrangement, we will commence a review of the existing practice of temporarily holding refunds pending the finalisation of credit income tax amendments.

Edit, scanning and exception checks

5.49 During this review the Inspector-General observed that there was a process carried out by the Operations area of the Tax Office to check that a refund amount shown on a paper-lodged GST return had been shown in whole dollars only. The Tax Office has advised that, during 2003-04, this process was applied to all paper-lodged refund claims which exceeded a certain small dollar value. This amounted to 45 per cent of all paper refund BASs.49 This review found that this process caused some GST refunds to be unnecessarily delayed — for up to several days in some cases.

5.50 The Inspector-General recommends that the Tax Office continue to explore alternative ways of dealing with this dollar and cents problem. The current procedure results in a large number of refunds being taken off-line from automatic processing procedures where only a very small percentage of those refunds will have the problem which the procedure is designed to detect.

5.51 The Inspector-General also notes that the Tax Office has not advised taxpayers or tax practitioners that GST refunds lodged by paper could be delayed for this reason.

5.52 The Inspector-General further notes that refunds can be delayed due to the Tax Office detecting errors which taxpayers have made in completing their GST return. The Tax Office publishes a list of certain errors it is finding in BASs in its Activity Statement Updates. However, these lists do not contain a complete listing of typical errors. The Tax Office has not published a stand-alone guide to provide taxpayers with guidance on the typical errors which are being made and how to prevent them.

5.53 The above observations lead to the following recommendations.

Subsidiary recommendation 4

The Inspector-General recommends that the Tax Office address the problem of ensuring that amounts shown on paper-lodged GST returns have been shown in whole dollars only in a way which does not involve large number of refunds being taken off-line from automatic processes.

Subsidiary recommendation 5

The Inspector-General recommends that the Tax Office publish on a regular basis, comprehensive lists of clerical errors commonly made on a BAS which could delay a GST refund.

Tax Office response to subsidiary recommendation 4

5.54 The Tax Office agrees with this recommendation and will further review the existing risk assessment processes.

5.55 With a view to reducing the number of activity statements taken off-line, we have already made a number of changes to reduce the incidence of scanning errors and in particular, errors relating to the inclusion of 'cents'.

5.56 We have made adjustments to the format of the activity statements:

  • Earlier versions contained the words 'Show whole dollars only'. This instruction now reads 'Show whole dollars only (do not show cents)'.
  • We have changed the format of the amount fields on the form to show crossed-out cent boxes. This indicates at each label that it is not appropriate to include cents.
  • We have adjusted our templates to shift the placement of fold lines on the form to ensure they do not fall across labels, causing misreads during the scanning process.

5.57 In addition to the physical form changes we take a proactive approach to client education. We publish details of common errors, including instructions not to include cents, in the Activity Statement Update bulletin. We are looking to improve our current broad-based approach to addressing errors, by implementing a model which identifies recurring mistakes and tailors our education response. In cases where adjustments are necessary due to recurring incidences of including cents on the activity statement, direct contact will be made with the client.

5.58 In addition, we will review the existing risk assessment process with an end to end focus to determine whether the material risk presented by the taxpayer recording cents on the activity statement is addressed by alternative controls in our processing systems.

Tax Office response to subsidiary recommendation 5

5.59 The Tax Office agrees with this recommendation and has been publishing information about common errors made by taxpayers in the GST system, including BAS errors. These lists have regularly featured in Activity Statement Update. Paper copies of Activity Statement Update accompany the quarterly BAS sent to taxpayers and electronic copies are placed on the ATO internet site.

5.60 The Tax Office will now publish a regular comprehensive list of BAS errors on an on-going basis via the Business and Tax Agent Portals and the ATO internet site. Activity Statement Update will include references to these information sources.

Verification activities which may delay a GST refund

5.61 Concerns raised by taxpayers in connection with delays caused by verification activities undertaken by the Tax Office on GST refund claims fall into the following two broad categories:

  • concerns relating to verification activities which are in the nature of desktop checks on the relevant refund claims; and
  • concerns relating to more detailed audit activities conducted on GST refund claims, for example, those which have involved a field visit to the relevant taxpayers.

5.62 Each of these concerns is discussed below.

Desktop verification procedures

Nature of concerns raised

5.63 A major concern of taxpayers raised in numerous submissions made to this review was that desktop verification processes, being the processes carried out by the various areas to which GST refunds were streamed after being flagged by the RRE, took too long to complete.

5.64 Submissions received from many taxpayers and their advisers noted that delays in processing GST refunds that were referred to verification areas for desktop checks affected particular industries — particularly exporters, property developers and taxpayers who made GST free supplies, such as those involving food or health services.

5.65 At one consultation meeting conducted with practitioners, several practitioners who were tax agents for pharmacies noted that these taxpayers regularly had their refunds delayed for lengthy periods whilst undergoing verification activities and that almost every second BAS was queried.

5.66 These practitioners commented that this has meant that some pharmacies are actually switching to a quarterly lodgment basis even though it means they are waiting considerably longer to claim their refunds. The reasoning behind the change is that, where a reviewed BAS has not been processed before another credit BAS arrives, the Tax Office 'stops' not only the new refund, but also the existing refund. This flow-on effect is a major problem and taxpayers have worked out they can circumvent the problem by allowing the Tax Office three months to release the first BAS refund before lodging the next one.

5.67 Submissions also noted that a refund could be especially delayed by these verification procedures where there had been a one-off transaction by a taxpayer — such as a large capital purchase. One submission by a professional association supplied a number of examples of this kind of case where the first contact was made by Tax Office verification staff with the relevant taxpayer or their agent only after four to five weeks had elapsed from the date of lodgment of the relevant BAS.50

Performance of verification areas in processing GST refunds

5.68 The Tax Office does not routinely prepare management information reports which indicate the turnaround times for GST refunds processed by verification areas conducting desktop reviews. The method by which each such verification area monitors its performance in the processing of GST refunds is by way of certain weekly reports. In the case of one verification area (ILEC) these weekly reports were only introduced during the course of this review.

5.69 During this review, the Tax Office was asked to prepare data for the 2003-04 year which indicated the length of time refunds that were subject to this form of verification spent in both the Tax Office overall and within each verification area prior to being paid.

5.70 The Tax Office was only able to provide data indicating the length of time such refunds spent within each verification area. This is not an ideal indicator of the total length of time which GST refunds that are subject to verification actually spend in the Tax Office. This is because such a refund will spend a minimum of one day in the Operations area of the Tax Office prior to being allocated to a verification area. It will spend a further one day in the Operations area after leaving the relevant verification area before it is approved for payment. The operation of the banking system will then mean that it will take an additional two days before the refund is actually paid into the taxpayer's bank account.

5.71 The data supplied by the Tax Office, while not an ideal method of assessing Tax Office performance on the processing of refunds that have been subject to verification, nevertheless provides some indication of the extent to which refunds which are streamed to verification areas are being processed by the Tax Office on a timely basis.

5.72 The first set of data supplied by the Tax Office was the median number of days that a GST refund case that is delayed for more than 14 days will spend in each relevant verification area. This data was as follows:

CVCs — 30 days

ILEC — 27 days

GCS — 21 days

General Field — 67 days51

5.73 The other data supplied by the Tax Office was material which recorded for each verification area, and also for particular types of verification cases within the CVC and ILEC verification areas, the number and dollar value of refunds that were processed in that area during intervals of 0-4 days, 5-9 days and then at various intervals up to 90-plus days.

5.74 The first 0-4 day period allows an assessment to be made of the extent to which each verification area is processing GST refunds under certain 'early release' procedures. These procedures allow refunds to be released within a very short time period without the relevant taxpayers being contacted.

5.75 The second time period of 5-9 days was selected because refunds that are processed by verification areas within this time period should, according to the Tax Office, generally be paid to the taxpayers within an elapsed time period of 14 days from the date of lodgment. This takes into account that it will take a minimum of three working days or five calendar days after a refund has been approved by a verification area for it to be processed by the Operations area of the Tax Office and the banking system and then paid into a taxpayer's account.

Processing performance for small business taxpayer refunds and other refunds processed by CVCs

5.76 The table below, provided by the Tax Office, indicates the degree to which refund cases processed within CVCs during 2003-04 were processed in that area within designated time intervals.

Table 1: GST refund cases referred to CVCs in 2003-04
Days Cases Per cent Refund $ Per cent
0-4 24,747 34.18 -2,306,995,785 45.90
5-9 10,989 15.18 -952,784,590 18.96
Sub-total 35,736 49.36 -3,259,780,375 64.86
10-14 7,520 10.38 -556,285,751 11.07
15-29 16,777 23.17 -868,993,096 17.29
30-59 10,018 13.83 -309,482,988 6.16
60-89 1,325 1.83 -29,134,573 0.58
90-plus 1,036 1.43 -2,257,405 0.04
Total 72,412 100.00 -5,025,934,188 100.00

5.77 The above table reveals that, during 2003-04, approximately 49 per cent of all CVC GST refund cases in terms of number and 65 per cent in terms of dollar value were processed within the CVC area in nine calendar days. This means that approximately 51 per cent of GST refunds in terms of number and 35 per cent in terms of dollar value were not paid to taxpayers within 14 days.

5.78 However, the table does indicate that, of the cases that have been processed within nine days, most of these (being approximately 34 per cent of the total number of CVC cases, and 46 per cent of the total dollar value of CVC cases) were processed within 0-4 days. This suggests that, during 2003-04, the early release processes adopted by the Tax Office may have accounted for a significant number of the CVC cases that were finalised within nine days in the CVC area itself (or 14 days in the Tax Office overall).

Inspector-General comment

5.79 The Tax Office's failure to process 51 per cent of the refunds that were subject to verification processes within the CVC area within nine days during the 2003-04 year appears, at first glance, not to be an acceptable performance result. However, there may be mitigating explanations for this outcome.

5.80 This statistic would, for example, be acceptable if, as the Tax Office has estimated in material provided to the Inspector-General, it is correct to say that at any single point in time approximately 50 per cent of the refunds that are delayed within the Tax Office are held up because the Tax Office is waiting for the taxpayer to provide relevant information.52

5.81 This statistic may also be acceptable on the basis of the amount of permanent tax adjustments that arise from these delayed refunds.

5.82 On the other hand, during this review, taxpayers and tax practitioners stated that refunds could be delayed while they were transferred between CVC centres in different states or within a CVC. Taxpayers and practitioners were aware of transfers involving different CVCs because, in some cases, they were asked to provide relevant tax invoices and other documentation supporting the refund claim to both the original CVC and then again to a different CVC. GST refund cases examined by the Inspector-General also indicated that transfers of cases between CVCs or between CVCs and other verification areas of the Tax Office could be a reason for a GST refund being unduly delayed. Particular examples of cases where this occurred were cases involving financial supplies.

5.83 The Tax Office's present systems are such that it can only speculate on possible causes for these refunds being delayed.

Changes introduced in 2004

5.84 The above table indicates the performance achieved by CVCs during the 2003-04 year. The Tax Office has advised the Inspector-General that the combined effects of new streamlined procedures introduced in early 2004 and a new override process introduced from July 2004 have meant that the performance times for CVC cases have improved from those indicated in the above table.

5.85 For example, the Tax Office has advised that, prior to the adoption of the proposed new override process, 46 per cent of CVC cases were being finalised via the early release of refund process as at August 2004.53

5.86 However, the Inspector-General notes that an increase in the rate at which cases are released under early release procedures will automatically improve the percentage of cases that are processed within nine days, without necessarily resulting in an improvement in the turnaround times for those cases that are subject to more detailed verification procedures involving contact with the taxpayer.

5.87 The Tax Office has also advised that, as at September 2004, 1,950 small business taxpayers that would normally have their refunds verified by a CVC have been subject to the above override process. The Tax Office anticipates that this will account for over 5,000 refunds. The Tax Office also expects that the number of overrides for small businesses will increase in future months.

5.88 The Tax Office further advises that its Strategic Risk Management area is also currently identifying public listed companies in its small business segment to determine the extent to which the override will apply for these entities. The cases selected will use the same criteria as are being used for large enterprise taxpayers, as discussed earlier in this report.54

Tax Office reporting processes for delayed refunds

5.89 During fieldwork in relation to this review, the Inspector-General observed that the Tax Office prepares weekly performance reports on cases held within CVCs. These reports provide considerable detail, including the aged status and dollar values of such cases.

5.90 However, these reports have a primary focus on the throughput of cases managed within CVCs rather than analysing why refunds have been delayed. They do not, for example, reveal the nature or size of the potential adjustments associated with the refunds which have been delayed.

5.91 Also, although these weekly reporting processes provide information on the throughput of refund cases, they have not operated to manage the throughput of all such cases.

5.92 Firstly, these weekly reports only provide aged status and other details for cases that have been held within the CVC for more than 10 days. These reports therefore do not provide details for CVC cases which have been in CVCs for 10 days or less.

5.93 Secondly, these weekly reporting processes do not immediately detect that certain GST refunds have been delayed for more than 14 days since lodgment of the relevant BAS.

5.94 These reporting processes only detect GST refund cases which have been delayed for more than 10 days from the time that they have actually reached a CVC. They do not detect and start to monitor GST refund cases which, at some time before this 10-day period expires, have become cases which have spent more than 14 days in the Tax Office due to the time they have already spent in the Operations processing area.

Processing performance for large enterprise taxpayer refunds

Processing performance for ILEC GST refunds overall

5.95 The table below, provided by the Tax Office, indicates the degree to which refund cases processed within ILEC during 2003-04 were processed in that area within designated time intervals.

Table 2: GST refund cases referred to ILEC in 2003-04
Days Cases Per cent Refund $ Per cent
0-4 1,635 42.17 -4,514,953,931 61.39
5-9 743 19.17 -1,938,669,988 26.36
Sub-total 2378 61.34 -6,453,623,919 87.75
10-14 385 9.93 -368,259,401 5.01
15-29 623 16.07 -386,587,034 5.26
30-59 323 8.33 -80,371,144 1.09
60-89 78 2.01 -18,880,494 0.25
90-plus 90 2.32 -47,101,413 0.64
Total 3,877 100.00 -7,354,823,405 100.00

5.96 This table indicates that, during 2003-04, around 42 per cent of ILEC refund cases in terms of numbers and around 61 per cent of such cases in terms of dollar value were processed within ILEC within 0-4 days. This means that at least 58 per cent of the total number of ILEC-held refunds and 39 per cent of the total dollar value of such refunds were not processed under ILEC's early release procedures.

5.97 At the end of nine days, 61 per cent of ILEC refund cases, in terms of number, and 88 per cent, in terms of dollar value, had been processed within ILEC. However, this still means that 39 per cent in terms of the number of held refunds and 12 per cent in terms of the dollar value of such refunds were not paid in the 14-day calendar day period after lodgment of a BAS.

Inspector-General comment

5.98 The Tax Office's processing performance for GST refunds involving large enterprise taxpayers is better than that achieved for refunds involving small business taxpayers. For example, for ILEC taxpayers the Tax Office has processed 42 per cent of GST refunds in 0-4 days in terms of number and 61 per cent of such refunds in terms of dollar value. For taxpayers whose refunds were reviewed by CVCs, the corresponding percentages were 34 per cent and 46 per cent respectively.

5.99 However, for large enterprise taxpayers 39 per cent of the refunds in terms of number and 12 per cent of such cases in terms of dollar value were not processed by the end of nine days within ILEC and 14 days within the Tax Office overall.

5.100 As with CVCs, this may or may not be an acceptable performance result, depending on the causes of these delays.

5.101 The incorrect referral of ILEC refund cases to other areas of the Tax Office prior to being sent to the correct area within ILEC is one cause of GST refunds being delayed. The Inspector-General observed this problem particularly in relation to refund cases involving financial supplies. This is discussed in further detail below. However, this problem was also observed for non-financial supplies GST refund cases.

5.102 Under its present systems the Tax Office does not know the nature and extent of causes of GST refunds being delayed, as it does not routinely track the actual causes of refund delays within ILEC.

5.103 The Inspector-General also notes that the sum of the net tax adjustments arising from refunds that were checked by ILEC, being $14.5 million, was considerably less than that which was achieved by CVCs. This figure suggests that the process of checking the vast majority of GST refunds for large enterprise taxpayers before they are paid, and delaying 39 per cent of the number of such refunds for more than 14 days, may not be justified on a risk management versus cash flow consequence basis.

Processing performance within different segments of ILEC

5.104 ILEC is divided into five industry segments and two general segments. The five industry segments deal with large taxpayers engaged in the following industries:

  • resources and energy;
  • technology, information, communications and entertainment ('TICE');
  • property and construction;
  • general insurance; and
  • financial services.

5.105 The two general segments are Cap North and Cap South. Cap North deals with all ILEC taxpayers in NSW and Queensland who do not fall within one of the above five specific industry segments. Cap South deals with all ILEC taxpayers who are in the other Australian states who do not fall into one of the above five industry segments.

5.106 For the financial services segment, one cause of refund delays is the incorrect allocation of ILEC GST refund cases to other areas within ILEC or to a CVC prior to being forwarded to the financial services segment of ILEC. This can occur because ILEC is responsible for processing the GST refunds for all financial services taxpayers. However, many financial services refunds will initially be allocated to a CVC. They will then be referred to the ILEC area if they are considered to be of a complex nature. Some financial services cases will also be allocated to another area of ILEC before they are identified as cases which should be dealt with by the financial services segment.

5.107 A number of the financial services cases observed during fieldwork conducted during this review had been reallocated five to twelve times, either within ILEC or between CVCs and ILEC, before being actioned by a tax officer.

5.108 The Inspector-General notes that the level of complexity of GST refunds involving financial services taxpayers may not be significantly greater than that which arises for other types of taxpayers, such as, for example, those involved in property transactions. Complexity may therefore not be a valid reason for the special processing structure which the Tax Office has adopted for financial supplies GST refund cases.

5.109 The Inspector-General also notes that, despite the high volume of GST refunds involving financial services taxpayers that have been subject to examination by ILEC staff, few adjustments have actually been made. The Tax Office has provided figures to the Inspector-General which indicate that, of the $14.5 million of net GST adjustments that were made as a result of the refund verification work performed by ILEC, $2.7 million related to taxpayers whose refunds were examined by the financial services segment.55

Inspector-General observations on ILEC's processing performance for GST refunds overall

5.110 The Inspector-General notes that the Tax Office has introduced an override arrangement for certain ILEC taxpayers. However certain ILEC taxpayers may still be subject to the existing pre-refund checking processes in ILEC. If this is the case, then the Tax Office needs to ensure that refunds that will continue to be checked by its ILEC area are processed on a timely basis. These procedures should include a focus on the reasons why refunds are being delayed within ILEC overall and within each ILEC segment, particularly within the financial services segment.

Processing performance for large Government and Community Sector taxpayer refunds

5.111 The GCS branch of the Tax Office is responsible for managing GST compliance for large taxpayers (that is, generally taxpayers with an annual turnover of $100 million or more) in the government, health, education and non-profit sectors.

5.112 Material provided to staff of the Inspector-General indicates that, for the 2003-04 financial year, there were 1,641 GST refund cases, representing approximately $7.16 billion in dollar value, referred to the GCS branch as a result of the RRE processes. Of these, 46 cases (or 3 per cent of cases), representing a total GST refund value of $45 million, were not processed within 14 days of lodgment.

5.113 The fieldwork conducted by staff of the Inspector-General indicated that one cause of long refund delays for certain GCS taxpayers is the allocation processes used by the Tax Office to determine whether a case should be referred to the GCS branch.

5.114 A GST refund case will be allocated to the GCS branch if the taxpayer's record within the Tax Office has both a large enterprise client (LEC) indicator on its account and an Australian and New Zealand Standard Industry Classification (ANZSIC) code which denotes that the taxpayer falls into one of the industry segments handled by the GCS branch.

5.115 The ANZSIC code which is recorded for each taxpayer is generally that which the taxpayer has nominated in its initial application for GST registration. This ANZSIC indicator may not therefore be a correct reflection of whether the taxpayer is the type of taxpayer which the Tax Office regards as a GCS taxpayer.

5.116 Of the 22 GCS branch-delayed GST refund cases examined by staff of the Inspector-General, 11 involved delays that were due to these allocation processes. Of these 11 cases, five involved large private sector clients with large refund amounts and involved delays ranging from five to 28 days.

5.117 The Inspector-General notes that the Tax Office will introduce more streamlined processes for GST refunds involving government entities under proposed new override processes. A certain number of taxpayers' GST refunds may still be reviewed by the GCS branch. For these taxpayers, the Inspector-General suggests that the Tax Office consider introducing a separate 'GCS' indicator similar to the existing LEC indicator so that any large GST refund cases that are to be examined by the GCS branch are streamed directly to that branch for examination.

Tax Office response made during the review

5.118 The Tax Office has advised that it is including in its information technology plans the introduction of a specific Government and Community Sector indicator that would operate in addition to the taxpayer-provided ANZSIC code.56

Verification procedures for GST refunds involving a detailed audit

Nature of concerns raised

5.119 During the course of this review, a number of taxpayers raised concerns about the Tax Office's policies for withholding GST refunds during the period that a detailed audit is being conducted by the Tax Office.

5.120 These concerns include the following:

  • all GST refunds due to a taxpayer could be delayed under this process for periods of up to 12 months or more, even where the taxpayer was engaged in the type of business which would ordinarily always give rise to GST refunds, such as an export business, or in a business which would give rise to very large refunds at certain times, such as a property developer;
  • the above delays in refunds could create very serious cashflow difficulties for the taxpayers concerned;
  • refunds could be withheld under this process even where the underlying issue had been previously resolved by the Tax Office in a review conducted a number of years earlier;
  • the audit process could be delayed by Tax Office staff requesting information already provided to the Tax Office;
  • the withholding of a refund in such cases prejudged the result of the audit;
  • the Tax Office has no legal power to withhold GST refunds;
  • a decision to stop refunds in these circumstances will not be reviewed internally by the Tax Office; and
  • only legal action is available to a taxpayer seeking the release of refunds prior to the conclusion of an audit.

5.121 The Inspector-General has not examined the nature of activities conducted by the Tax Office during detailed audits involving GST refunds, nor the above taxpayer concerns with these activities, as these subjects extend beyond the terms of reference of this review.

5.122 However, the subject of the ATO's audit practices generally may be the subject of further review by the Inspector-General in accordance with the priorities of his work program.

Tax Office performance standards

5.123 The Tax Office does not have separate performance standards for the processing of GST refunds and BAS refunds. However, it does have a separate service standard for the processing of activity statement refunds. As indicated earlier, activity statement refunds can involve the refund of taxes other than the GST.

5.124 The service standards for activity statement refunds are part of the Taxpayers' Charter. These standards are as follows:

  • The Tax Office will aim to process activity statement refunds within 14 days of receipt in the Tax Office.57
  • Where the Tax Office requests additional information to complete the processing of an activity statement refund, it will aim to process that refund within 14 days of the receipt of that additional information.58

5.125 The Tax Office has advised that it currently measures its effectiveness in processing GST refunds based on a standard of 92 per cent of activity statement refunds being issued within 14 days.59

International performance standards for GST refunds

5.126 The Tax Office advises that it has been monitoring the performance of other revenue agencies in the processing of GST refunds through their annual reporting documents (where they are in English and relevant). The four countries the Tax Office has examined are the United Kingdom (UK), Ireland, New Zealand (NZ) and Canada.60

5.127 The Tax Office has prepared the following table summarizing performance for 2002-03 for these countries compared with Australian performance for 2003-04.61

Table 3: Comparison of Australian GST refund performance in 2003-04 with 2002-03 refund performance of certain other countries
Country Performance standard Standard achieved
Australia 92% in 14 days 94.4%
UK 90% in 14 days N/A
UK 90.3% in 30 days by value 90.8%
Ireland 85% in 10 days 80%
Ireland 100% in 30 days 93%
NZ 100% in 21 days 97.9%
Canada 100% in 21 days 96%

5.128 The Tax Office notes that in comparing the above figures the following points should be noted:

  • the Australian standard applies to returns whether verified or not;
  • the UK standard only applies to correct returns. Interest only applies after 30 working days;
  • the New Zealand standard is based on 'non-queried' returns. It does not include refunds where manual intervention is required; and
  • Canada reviews 41 per cent of refund returns, compared to the Tax Office's review of approximately 4 per cent of refund returns. Canada formally audits 5 per cent of refund returns.62

5.129 The Tax Office considers that, in the light of the above table, its performance in processing BAS refunds:

'is likely to be at world's best practice. From this comparison it is clear that the Tax Office compares favourably.'63

5.130 However, the Inspector-General notes that the above international comparison of BAS refund processing times must be viewed with some caution, bearing in mind that differences exist in the administration and policy of goods and services taxes in the above countries.

5.131 A number of the differences in administration are referred to in Appendix 3 of the ANAO's report in March 2004 on activity statement processing.64 For example, this report notes that, at the time of the report, in Canada, lodgment by way of the internet was not available and in the United Kingdom all VAT returns, including those that were submitted electronically, were printed out when received and keyed in by staff.

Time frames achieved

5.132 The Tax Office has reported that, for the 2003-04 year, it exceeded the above standard of 92 per cent within 14 days in each month. It states that during this year it processed 94.4 per cent of activity statement refunds (in terms of the number of cases) within 14 days.65

5.133 The Tax Office measures its achievement against the 14-day performance standard on the basis of the number of refunds which attract payment of delayed refund interest against the total number of activity statement refunds processed.66

5.134 Delayed refund interest is generally calculated from 14 days after the date of lodgment. However, in certain cases, such as where a BAS is incomplete or incorrect, bank account details have not been provided, there are outstanding BASs or where the Tax Office has requested further or fuller GST returns from the taxpayer, this interest is payable only from the date which is 14 days after the relevant outstanding information is provided.

5.135 The Tax Office calculation of the number of activity statement refunds that have not been paid within 14 days does not consist of all cases where a refund has been delayed for more than 14 days after lodgment. Instead, it consists of those cases which have been delayed in circumstances which attract delayed refund interest.

5.136 The Tax Office has advised that it processed 1.89 million BAS refunds during 2003-04 and paid out $22.219 billion of net BAS refunds.67

5.137 As noted above, the 94.4 per cent performance result achieved by the Tax Office does not specifically apply to GST refunds, but to activity statement refunds generally. The Tax Office is not able to provide a break-up of this result for GST refunds specifically.

5.138 However, if this same result does apply to GST refunds, the material supplied by the Tax Office shows that, during 2003-04, approximately 5.6 per cent of 1.89 million BAS refunds, or approximately 106,000 BAS refunds, were not processed and paid within 14 days of all relevant information being provided to the Tax Office.

Inspector-General comments

5.139 The current performance standard for GST refunds is based on a view that GST refunds, BAS refunds and activity statement refunds are all processed in the same time frames. It is also based on a view that the Tax Office's performance in processing GST refunds should be judged on the number of cases where it was required to pay delayed refund interest, rather than the number of cases where a refund was not paid within 14 days.

5.140 The current performance standard for GST refunds also focuses on the number of activity statement refunds processed within a certain time period. It does not refer to the dollar value of those refunds.

5.141 The principal means by which the Tax Office monitors the dollar value of delayed refunds cases is through the measures adopted to monitor the amount of delayed refund interest paid.

5.142 During 2003-04, the performance standard used by the Tax Office has not given a transparent view of the nature of GST refunds that were not paid within 14 days of lodgment. As indicated previously in this report, 4.3 per cent of the total number of GST refunds were stopped for manual checking. These involved approximately $20 billion of refunds in dollar value.

5.143 Of this $20 billion, at least $1.8 billion (approximately)68 were GST refunds for small business taxpayers that were not paid within 14 days of lodgment. At least $0.9 billion (approximately)69 were GST refunds for large enterprise taxpayers that were not paid within this time period. A further $45 million were GST refunds that were not paid within 14 days of lodgment for government and community sector taxpayers. An additional $850 million were GST refunds referred for a GST field audit or to the Tax Office's Serious Non Compliance area and were therefore unlikely to have been paid within 14 days of lodgment.70

5.144 These figures indicate that, during 2003-04, at least 16 per cent in dollar value terms (being $3.6 billion out of $22 billion) of GST refunds was not paid within 14 days of lodgment. These figures do not include GST refunds held for more than 14 days since lodgment due to the absence of bank account details or outstanding BASs.

5.145 The above comments therefore lead to the following recommendation.

Subsidiary recommendation 6

The Inspector-General recommends that the Tax Office supplement its existing performance standard for activity statement refunds processing, which is based on the number of refunds processed, by regularly publishing supplementary management information which indicates the average dollar value of refunds held for more than 14 days after lodgment for either verification purposes or for other reasons.

Tax Office response

5.146 The ATO agrees with this recommendation. This information will be provided with our performance statistics for our 2004-05 outcomes.

End to end management of GST refunds

5.147 During the course of this review, the Inspector-General has found that each area of the Tax Office responsible for dealing with GST refunds (other than the area which deals with the processing of GST refunds after they have been cleared for payment) had procedures to ensure that such refunds were actively managed within the relevant area and therefore were not subject to undue delay.

5.148 However, the Inspector-General has found that these procedures are defective in a number of respects. The principal defect is that that they only operate to monitor the length of time a GST refund spends within each specific area of the Tax Office. There is no overall process which exists to monitor the length of time from lodgment to the actual time of payment.

5.149 These processes have meant that refunds which have been delayed within one area of the Tax Office have, for example, not been treated as cases requiring priority handling when they are subject to other subsequent stages of processing within the Tax Office. Essentially, these processes mean that the Tax Office is not actively managing its processing of GST refunds on an 'end to end' basis.

5.150 The Tax Office does not prepare historical reports which indicate why, for example, in the 2003-04 year, approximately 106,000 BAS refunds attracted delayed refund interest.71 Similarly, there are no reports available which indicate why, during 2003-04, other refunds which did not attract delayed refund interest were delayed for more than 14 days after the date of lodgment of the relevant BAS. The number of these other refunds may add considerably to the 106,000 figure, given that the Tax Office has estimated that, at any single point in time, approximately 85 per cent of BAS refunds that are being held by the Tax Office for more than 14 days involve refunds which do not attract delayed refund interest.72

5.151 The Tax Office does prepare certain snapshot reports. These reports do not allow determination of the predominant causes of refunds being delayed over an elapsed period of time, such as the 2003-04 year, and of what steps are necessary to reduce these causes of delay.

5.152 The above observations lead to the following recommendation.

Key recommendation 2

The Inspector-General recommends that the Tax Office establish 'whole of office' systems which measure the total elapsed time for the payment of GST refunds.

Tax Office response

5.153 The Tax Office agrees with this recommendation. By June 2005, the Tax Office will be in a position to regularly measure the elapsed time for activity statement refunds, and will address remaining systemic and procedural issues which might currently impede a 'whole of office' view for refunds. The Refund Integrity Steering Committee and Refund Business Management Group will oversee this work, and will consult with the Inspector-General's office in resolving issues.

GST refunds delayed within the 14-day period

5.154 The Tax Office also does not currently prepare reports which show actual GST refunds achievement within the period of 14 days from lodgment, for example, how many refunds are processed within 0-4 days or 5-9 days etc of lodgment.

5.155 A number of submissions made to this review suggested that there are a very significant number of GST refunds which are delayed beyond the time they should reasonably take to process. In particular, they assert that many refunds are paid just inside a 14 -day period.

5.156 In the absence of any Tax Office reports being prepared to monitor the number of GST refunds which are held for time periods of less than 14 days, the Tax Office does not have systems in place to enable comment on the validity of these claims, or to facilitate appropriate steps being taken to address these concerns.

5.157 Taxpayers and tax practitioners have further stated that they do not understand why, in a self assessment environment, the Tax Office uses the 14-day period as its benchmark for the payment of GST refunds.

5.158 It was submitted that, at least for routine refunds73 which are not taken off-line for verification activity, a shorter time frame was more appropriate.

5.159 A shorter time frame was considered to be especially applicable to electronically lodged activity statements.

5.160 As discussed earlier in this report, the Tax Office's performance statistics for electronically lodged returns for the 2003-04 year suggest that it is appropriate to advise taxpayers that they can expect routine refunds lodged via electronic means to be in their bank accounts within six working days of lodgment (or eight calendar days, including a weekend).

5.161 The above comments lead to the following recommendation.

Subsidiary recommendation 7

The Inspector-General recommends that the Tax Office introduce an eight day service standard for the processing of 'routine' electronically lodged GST refunds.

Tax Office response

5.162 The Tax Office agrees with this recommendation. The Tax Office undertakes to process all routine activity statements lodged electronically within eight days. Tax Office service standards will be amended to reflect this.

5.163 The Tax Office considers 'routine' activity statements to be those that go through processing without any manual intervention, including exception errors and verification activities.


35 Commissioner of Taxation, The Art of Tax Administration: Two Years On, Speech given to the 6th International Conference on Tax Administration: Challenges of Globalising Tax Systems, 15 April 2004, downloaded from www.ato.gov.au on 5 May 2004.

36 ATO Minute No IGT-GST01-2004 at paragraph 2.1 on p. 4.

37 This figure is derived from figures supplied by the Tax Office in ATO Minute No IGT-GST16- 2004.

38 ibid.

39 Commissioner of Taxation, Annual Report 2003-04 on p. 48.

40 Australian Taxation Office 'Why Activity Statement refunds are delayed or held', issue recorded in Issues Register of the ATO/Tax Practitioners' Forum, downloaded from www.ato.gov.au on 5 May 2004.

41 ATO Minute No IGT-GST06-2004 at paragraph 1.2 on p. 3.

42 ATO Minute No IGT-GST13-2004 at paragraph 2.4.

43 ATO Minute No IGT-GST13-2004 at p. 4 and ATO Minute No IGT-GST16-2004 at paragraph 5.1 on p. 6.

44 The Tax Office has advised that examples of cases where it will pay a GST refund by way of a cheque in favour of the taxpayer include cases where the taxpayer's religion precludes the operation of a bank account and where the taxpayer is either a foreign diplomat or a foreign government entity, and has no place of business or residence in Australia (ATO Minute No IGT-GST02-2004 at paragraph 4.4.3).

45 ATO Minute No IGT-GST02-2004 at paragraph 2.3 on p 4.

46 ATO Minute No IGT-GST06-2004 at paragraph 1.2 on p 3.

47 ATO Minute No IGT-GST11-2004 at paragraph 2.1 and 2.2 on p. 6.

48 These are contained in Operations Practice Note 2004/024 which forms part of attachment B to ATO Minute No IGT-GST06-2004.

49 ATO Minute No IGT-GST16-2004 at paragraph 2.3 on p. 3.

50 Taxation Institute of Australia, Review into the tax administration of GST refunds arising from the lodgment of business activity statements, Submission to Inspector-General of Taxation dated 8 June 2004.

51 ATO Minute No IGT-GST15-2004 at p. 4.

52 ATO Minute No IGT-GST03-2004, dated 7 June 2004, at paragraph 9.3.

53 ATO Minute No IGT-GST09-2004 at paragraph 5.4.2.3 on p. 10.

54 ATO Minute No IGT-GST11-2004 at p. 5.

55 ATO Minute No IGT-GST09-2004 at p. 6.

56 ATO Minute No.IGT-GST11-2004 at paragraph 6.2.

57 Australian Taxation Office, Our Service Standards, downloaded from www.ato.gov.au on 4 August 2004.

58 ATO Minute No.IGT-GST03-2004, dated 7 June 2004, at paragraph 7.2 on p. 9.

59 ibid at paragraph 7.4.

60 ibid at paragraph 10.2.

61 ibid at paragraph 10.3.

62 ibid at paragraphs 10.4 to 10.8.

63 ibid at paragraph 10.8, p. 11.

64 Australian National Audit Office, The Australian Taxation Office's Collection and Management of Activity Statement Information, Audit Report No. 33 2003/4 Performance Audit.

65 ATO Minute No IGT-GST18-2004 at p. 3.

66 ATO Minute No IGT-GST14-2004 at paragraph 11.1.

67 ATO Minute No IGT-GST09-2004 at p. 7.

68 This amount has been calculated as 35.14 per cent of $5,025,934,188 and is based on the figures shown in the table in paragraph 5.76 of this report.

69 This amount has been calculated as 12.25 per cent of $7,354,823,405 and is based on the figures shown in the table in paragraph 5.95 of this report.

70 ATO Minute No IGT-GST15-2004 at paragraph 9.1 on p. 6.

71 ATO Minute No IGT-GST18-2004 at paragraph 2.1 on p. 3.

72 ATO Minute No IGT-GST01-2004 at paragraph 1.1.

73 As discussed in paragraph 5.4 of this report, the term 'routine refunds' refers to refunds that are not stopped for verification or other reasons.