Skip to main content
Normal View

COMMITTEE OF PUBLIC ACCOUNTS debate -
Thursday, 10 Apr 2003

Vol. 1 No. 15

Annual Report of the Comptroller and Auditor General and Appropriation Accounts, 2001.

Vote 9 - Office of the Revenue Commissioners.

Mr. Frank Daly (Chairman, Revenue Commissioners) called and examined.

We are dealing with item No. 6, the annual report of the Comptroller and Auditor General and the Appropriation Accounts Vote 9 - Office of the Revenue Commissioners. Sections 4.1 to 4.6 and 4.8 and 4.9. are to be dealt with by the Revenue Commissioners. Section 4.10 is to be addressed also by the Department of Finance.

Members and witnesses' attention is drawn to the fact that as from 2 August 1998, section 10 of the Committees of the Houses of the Oireachtas (Compellability, Privileges and Immunities of Witnesses) Act 1997, grants certain rights to persons who are identified in the course of the committee's proceedings. These rights include the right to give evidence, the right to produce or send documents to the committee, the right to appear before the committee, either in person or through a representative, the right to make a written and oral submission, the right to request the committee to direct the attendance of witnesses and the production of documents and the right to cross-examine witnesses. These rights may only be exercised with the consent of the committee. Persons being invited before the committee are made aware of these rights and any persons identified in the course of proceedings who are not present may have to be made aware of these rights and provided with a transcript of the relevant part of the committee's proceedings if the committee considers it appropriate in the interests of justice. Notwithstanding this provision in legislation I remind Members of the long standing parliamentary practice to the effect that Members should not comment on, criticise or make charges against a person outside the House or an official, either by name or in such a way as to make him or her identifiable. Members are also reminded of the provisions in Standing Order 156 that the committee shall also refrain from inquiring into the merits of a policy or policies of the Government or a Minister of the Government or the merits of the objectives of such a policy or policies.

I invite Mr. Daly to introduce his officials to the committee.

I am accompanied by Mr. Christy Clayton, director general of revenue strategy and also Accountant General of the Revenue Commissioners. Mr. Paddy O'Shaughnessy is our principal officer and is Comptroller and Auditor General liaison officer. Mr. TomDowling is our administrative budget manager. Mr. Paddy Molloy is principal officer of our statistics branch and Ms Fionnuala Ryan is the assistant principal in that branch. Our press officer is in attendance in the public gallery.

I invite Mr. Hurley from the Department of Finance to introduce his colleagues.

Mr. David Hurley

I am a principal officer in the organisation management and training division dealing with Vote control. My colleagues from the budget and economic division are Mr. Dermot Mulligan, principal officer and Mr. Donal Murtagh, assistant principal.

Paragraphs 4.1 to 4.6 of the report of the Comptroller and Auditor General read:

Chapter 4 - Office of the Revenue Commissioners

4.1 Revenue Account

Basis for Audit

An account showing all revenue received and paid over to the Exchequer by the Revenue Commissioners is furnished to me annually. I am required under Section 3 of the Comptroller and Auditor General (Amendment) Act, 1993 to carry out such examinations of this account as I consider appropriate in order to satisfy myself as to its completeness and accuracy and to report to Dáil Éireann on the results of my examinations. The results of my examinations have been generally satisfactory.

I am also required under Section 3 of the Comptroller and Auditor General (Amendment) Act, 1993 to carry out such examinations as I consider appropriate in order to ascertain whether systems, procedures and practices have been established that are adequate to secure an effective check on the assessment, collection and proper allocation of the revenue of the State and to satisfy myself that the manner in which they are being employed and applied is adequate.

Revenue Collected

Revenue collected under its main headings in 2001 is shown in Table 1.

Table 1 - Revenue Collected

Gross Receipts €m

Repayments €m

Net Receipts €m

2000 Net Receipts €m

Income Tax

10,549

1,231

9,318

9,126

Value Added Tax

10,548

2,641

7,907

7,467

Excise

4,382

169

4,213

4,424

Corporation Tax

4,623

479

4,144

3,885

Stamps

1,257

34

1,223

1,091

Customs

171

6

165

207

Capital Acquisitions Tax

174

6

168

224

Capital Gains Tax

890

14

876

773

Residential Property Tax

2

1

1

1

Total

32,596

4,581

28,015

27,198

Of the net receipts of €28,015m, a total of €168m was paid during 2001 under Section 3 of the Appropriation Act, 1999 from the proceeds of tobacco excise to the Vote for Health and Children and €27,916m was paid into the Exchequer. As a result, there was a balance of €131m prepaid to the Exchequer at year end compared to a balance of €62m prepaid at the end of the previous year. As the final lodgment to the Exchequer at year end is required to be made on 31 December, before final reconciliations for each taxhead can be completed, there is necessarily an element of estimation which can result in over or under lodgments by Revenue to the Exchequer.

4.2 Outstanding Taxes and PRSI

Table 2 was prepared on the basis of information furnished by the Revenue Commissioners and reflects the activities and transactions in the twelve month period ended 31 May 2002 - the latest date for which data was available at the time of finalising my Report. Table 3 sets out an aged analysis of the balance outstanding at 31 May 2002.

Table 2 - Outstanding Taxes and Levies

Balance at 31 May 2001 €m

Tax or Levy

Charges/Estimates Raised €m

Paid €m

Balance at 31 May 2002 €m

Estimate of amount likely to be collected €m

160

VAT (Declared Liabilities Net of Repayments)

7,120

7,161

119

104

210

VAT (Estimates)

38

41

207

113

94

PAYE (Declared Liabilities)

12,136

12,061

169

118

28

PAYE (Estimates)

169

170

27

20

98

PRSI (Declared Liabilities)

8,035

7,936

198

141

20

PRSI (Estimates)

128

128

19

16

474

Income Tax (Excluding PAYE)

1,487

1,518

443

328

-

DIRT

214

214

-

-

291

Corporation Tax

4,123

4,225

189

126

93

Capital Gains Tax

910

892

111

92

17

Capital Acquisitions Tax

176

175

18

14

8

Abolished Taxes

1

1

8

-

1,493

Total

34,537

34,522

1,508

1,072

Table 3 - Aged Analysis of Debt at 31 May 2002

Tax

Total tax outstanding at 31 May 2002 €m

Amountsoutstandingfor period30/4/ - 31/12/01 €m

Amounts outstanding for 2000/01 €m

Due for periods 1990/91 to 1999/00 €m

Due for earlier periods €m

VAT

326

125

69

129

3

PAYE

196

58

40

78

20

PRSI

217

79

43

73

22

Income Tax

443

2

185

229

27

Corporation Tax

189

42

27

98

22

Capital Gains Tax

111

6

50

52

3

Capital Acquisitions Tax

18

18

-

-

-

Abolished Taxes

8

8

-

-

-

Total

1,508

338

414

659

97

The balance outstanding at 31 May 2002 of €1,508m is €15m greater than at the same point in 2001. It is estimated by Revenue that €1,072m or 71% of the total outstanding is likely to be eventually collected. This compares with an estimated collection ratio of 63% at May 2001. The estimation of the amount likely to be collected takes into account such factors as anticipated reductions of estimated amounts brought forward, the level of liquidations and business closures and historical business patterns. The movement of the tax year to coincide with the calendar year has caused some alteration in the manner of computation of debt ^ the most notable being the dramatic increases in the PAYE and PRSI arrears categories.

4.3 Revenue Audit Programme

Overall Audit Programme

In a self assessment system returns filed by compliant taxpayers are accepted as the basis for calculating tax liabilities. The validity of returns is established by the auditing of a selection of cases either through reviewing and seeking further verification of particular details or by the examination of documents and records at a taxpayer's premises. The majority of audits carried out by the Revenue Commissioners are specific to taxheads such as VAT or PAYE, but a significant number of comprehensive audits are also carried out. These may focus on all taxes but are primarily aimed at Income Tax, Corporation Tax and Capital Gains Tax.

In the course of my audit, a small representative sample of audit settlements were reviewed with satisfactory results. The outcome of the 2001 programme of Revenue audits is summarised in Table 4, which also includes 40 audits arising from investigation and anti-avoidance activity.

Table 4 - Revenue Audit Programme

Audit Type

2001

2000

No. of audits completed

Yield €m

No. of audits completed

Yield €m

Comprehensive

2,200

74.4

2,270

68.3

Value Added Tax

4,223

61.2

4,409

35.0

PAYE Employers

1,443

10.3

2,104

11.9

Relevant Contracts Tax (RCT)

383

3.0

352

1.7

Combined Fiduciary (VAT, PAYE and RCT)

626

11.8

670

6.0

Capital Acquisitions Tax

334

6.3

388

3.7

Verification Audits and Desk Reviews

7,107

37.2

6,126

10.2

Investigation Branch

30

0.4

4

0.3

Anti-Avoidance

10

3.9

7

2.4

’Pick-Me-Ups’

-

-

21

0.1

DIRT

47

-

37

220.0

Total

16,403

208.5

16,388

359.6

Comprehensive Audits

The selection of cases for comprehensive audit is made on the basis of such factors as screening of annual returns, re-audit of cases with previous undercharges, other information available to Revenue and random selection. Generally, a settlement is agreed following completion of the audit and any outstanding amount is paid. A number of settlements involve the restriction of losses which may be carried forward against future years' profits. Where an Inspector is unsuccessful in collecting the additional amount of tax and interest arising on audit adjustments, the amounts are referred to the Collector General for collection.

The outcome of the 2,200 comprehensive audits completed in 2001 is detailed in Table 5. The highest individual settlements were €1,490,990 for Income Tax and €5,966,902 for Corporation Tax. The overall yield of €74.4m includes interest charges of €13.1m and penalties of €7.9m.

Table 5 - Yield from Comprehensive Audits

Income Tax

Corporation Tax

Number ’000

Yield €m

Number ’000

Yield €m

Agreed Settlements

€1 to €12,697

561

4,325

176

1,563

€12,698 to €63,487

300

8,407

109

3,205

€63,488 to €126,974

59

5,353

34

2,905

Over €126,974

56

16,439

49

27,893

Other Settlement Activity

Returns accepted - no additional tax payable

577

-

236

-

Settled by restriction of losses carried forward to future years

25

104

9

3,422

Referred to Collector General for enforcement action

8

711

1

15

Totals

1,586

35,339

614

39,003

Random Audits

It is Revenue policy that 6% of cases selected for audit as part of the comprehensive, VAT and PAYE/PRSI audit programmes are selected randomly. The 6% policy would indicate a target of approximately 472 random audits for completion in 2001. In the event, 740 random audits were completed in 2001 consisting of 443 comprehensive (20%), 165 VAT (4%) and 132 PAYE/PRSI (9%). The returns of 510 taxpayers were accepted as originally submitted while additional liabilities of €3,381,890, including €1,033,281 in interest and penalties, were assessed in the other 230 cases.

4.4 Revenue Prosecution Activity

Prosecutions for Serious Tax Evasion

Under Revenue prosecution strategy, audit districts are required to forward cases to Investigation Branch for investigation with a view to criminal prosecution where there is prima facie evidence of serious revenue offences having been committed. These cases are further evaluated within the Branch before commencement of the very resource intensive criminal investigation work which can take several years before reaching the Courts. Convictions were obtained in all 4 of the cases decided in Court in 2001

· An accountant who was a director of a company which provided payroll services to film production companies, was convicted of submitting false PAYE returns and sentenced to twelve months imprisonment.

· An individual whose company provided security services was convicted of submitting incorrect Income Tax returns, delivering incorrect information and failing to keep proper records. A sentence of six months imprisonment and a fine of €1,905 was imposed. On appeal, the prison sentence was suspended and a further fine of €1,270 imposed.

· A construction company director was convicted of submitting incorrect VAT and PAYE returns and received a six months suspended sentence and was fined €2,750. The company was fined €7,618.

· An individual providing plumbing and heating services was convicted of submitting false VAT returns, obtaining VAT repayments by false pretences and failing to keep proper records. A three months prison sentence was imposed.

Of a total of 31 cases on hands at the end of 2001, 17 are still under investigation, 4 cases have been submitted to the DPP for consideration for prosecution, 6 are proceeding to prosecution and 4 cases have been closed. In one further case, the defendant pleaded guilty in the District Court in 2001. However, when the case came up for sentencing in April 2002, the District Justice refused jurisdiction and sent the case for trial in the Circuit Criminal Court. In June 2002, the Circuit Criminal Court gave leave to apply for a judicial review of the District Court decision.

Prosecution of Non-Filers

Taxpayers failing to submit returns of Income Tax and Corporation Tax normally receive a warning letter from the Revenue Solicitor. In the event that returns are still not submitted, legal proceedings are instituted. During 2001, 11,656 warning letters were issued and 1,101 cases (1,050 Income Tax and 51 Corporation Tax) were successfully prosecuted with fines totalling €1,045,354. Court orders were obtained in 40 of these cases which required the convicted person to submit all outstanding returns.

Penalties of €1,148,085 were imposed by Revenue on 476 employers who failed to make P35 employer returns on time. €372,255 of these penalties have been paid by 158 employers. Revenue are seeking to recover penalties of €347,220 from 139 cases by means of court proceedings. Penalties of €413,400 imposed on 173 cases have been withdrawn. The remaining six cases where penalties of €15,210 were imposed are under review as payment of the penalty is being offset from overpaid taxes.

4.5 Special Investigations

DIRT

In 2000, Revenue completed the DIRT 'look-back' audits of financial institutions and reported the results to the Committee of Public Accounts. Audits of all of the institutions have been completed with the exception of IIB Bank Limited and Guinness and Mahon (Ireland) Limited. Final liability in respect of these may depend on the outcome of investigations into the Ansbacher Accounts. Payments on account have been received from the two institutions. In 2001, 47 audits of financial institutions were completed to ensure continuing compliance with DIRT regulations.

Also in 2001, Revenue began to address the issue of the underlying tax on funds deposited in bogus non-resident accounts. The approach adopted by Revenue involved a deadline of 15 November 2001 for depositors to voluntarily disclose and pay any tax, interest and penalties. For those who availed of this, interest and penalties were capped at 100%, no prosecutions were taken and settlement details were not published. Under this arrangement, €227m was paid by 3,675 account holders in respect of 8,380 accounts. Of these, 599 account holders declared a nil liability. All returns are being checked for basic eligibility. In addition, samples have been selected both by the Underlying Tax Project Team and the various Tax Districts for review of the amounts disclosed.

Revenue are now seeking to identify bogus non-resident account holders who did not avail of the voluntary disclose and pay arrangements. Some 1,800 depositors identified during the DIRT look back audits have been referred to tax districts for investigation. The status of these cases at June 2002 is

· 108 have been settled with no liability

· 7 have settled with total liability of €352,724

· 166 have made submissions to Revenue and payments on account of €148,304 have been received from some of these

· 5 cases are being considered for prosecution

· the remaining cases are being researched further.

Revenue have applied for 18 High Court orders to obtain information from financial institutions relevant to a person's tax liability and 17 orders have been granted. It is expected that this investigation will continue for a number of years.

Offshore Investments via National Irish Bank

The investigation into individuals who invested in an offshore investment scheme operated by National Irish Bank is continuing. By August 2002, settlements were reached in 342 cases totalling €31m including interest and penalties of €17m. Of these cases, 90 were settled with no liability. In addition, payments on account totalling €7m have been received in respect of other unresolved cases. Seven cases have been referred for criminal investigation with a view to prosecution through the courts.

In 2001, National Irish Bank submitted to Revenue a list of 22 new cases. Of these, 15 involve relatively small sums and are thought unlikely to involve substantial tax evasion. Revenue have committed to investigate all additional cases as appropriate.

Ansbacher (Cayman) Limited

A special project team of eight investigators together with support staff is investigating the Ansbacher accounts. The team is investigating cases directly involving Ansbacher type arrangements as well as other cases involving offshore funds and deposits. At August 2002, a total of 300 cases were under investigation.

Investigations into five cases have been concluded as follows

· One case was settled with no liability

· One Ansbacher case was settled for €330,971, including interest and penalties of €190,461

· One case involving Ansbacher type arrangements was settled for €292,040, including interest and penalties of €169,580

· Two other cases involving offshore funds or deposits were settled for €279,342 (interest and penalties €144,996) and €250,138 (interest and penalties €177,399)

Payments on account totalling €17.23m have been received to date in 51 cases as follows

· €11.24m from 44 cases involving Anshbacher-type arrangements

· €5.99m from 7 cases involving offshore funds and deposits.

In July 2002, the report of the High Court Inspectors into Ansbacher (Cayman) Limited was published. Revenue are examining the report and comparing the information in it with the results to date of their own investigations. Revenue have also applied to the High Court for additional documents gathered by the Inspectors but not published in their report. This application is expected to be heard in November 2002.

Pick-Me-Up Schemes

Pick-Me-Up Schemes involved expenses for goods or services incurred by a political party being invoiced by the supplier to another trader who paid the supplier as a means of supporting the party. Such payments were not deductible for tax purposes, the VAT was not reclaimable and the invoices issued were not in accordance with legal requirements. The investigation has found that while some traders treated these payments correctly for tax purposes quite a large number did not. There were a total of 71 cases that apparently avoided tax by engaging in 'picking up' expenses which were proper to political parties. Of these, 15 have been mentioned at either the Flood or Moriarty tribunals and payments on account have been received from six of these. Of the remaining 56 cases, 42 have been settled. In 36 of these, no settlement exceeded €12,697, the total amount of such settlements being €144,233 including €79,913 in interest and penalties. Six of the settled cases involved payments of amounts greater than €12,697. The total yield from these was €326,491 including interest and penalties of €202,158.

Tribunals

Matters disclosed at the Moriarity and Flood Tribunals which suggest that tax evasion may have occurred are being investigated as they come to notice.

4.6 Write Offs

The Revenue Commissioners have furnished me with details of taxes written off during the year ended 31 December 2001. Details of the total amount written off and the distribution according to the grounds of write-off are shown in Table 6 and Table 7.

Table 6 - Taxes Written Off

Tax

2001 €’000

2000 €’000

Value Added Tax

29,476

42,895

PAYE

12,790

16,763

Corporation Tax

11,270

6,914

Income Tax

68,092

19,163

Other Taxes

6,401

2,232

PRSI

12,263

16,292

Total

140,292

104,259

Table 7 - Grounds of Write Off

2001

2000

No. of Cases

€’000

No. of Cases

€’000

Liquidation/Receivership/Bankruptcy

382

26,942

397

24,834

Ceased trading - no assets

578

16,945

1,432

41,875

Deceased and Estate Insolvent

52

1,631

144

3,275

Uneconomic to pursue

35,173

82,552

1,196

15,486

Unfounded Liability

37

830

34

582

Cannot be traced / Outside Jurisdiction

117

4,096

297

5,984

Compassionate Grounds

70

1,545

89

1,503

Uncollectable due to financial circumstances of taxpayer

243

5,660

382

10,114

Examinership

2

91

4

606

Totals

36,654

140,292

3,975

104,259

The write off in 2001 included the write off on an automated basis of 33,791 cases totalling €67m in respect of Income Tax, Corporation Tax and Capital Gains Tax. The larger cases were manually checked to confirm that the write off was justified. All amounts were pre 1991 and no amount was greater than €32,000. Cases under the control of the Criminal Assets Bureau, cases under investigation and potential Ansbacher cases were excluded from write off.

The Internal Audit Branch in Revenue undertakes an annual examination of a sample of cases written-off. The internal audit of 2001 write-offs in which 163 cases were examined has recently been completed. The audit did not include the automated write off. The results of the audit were generally satisfactory and no instances were found where tax was improperly written off.

I have examined a sample of cases representing over 4% of the value written off through a review of the procedures followed and of supporting reports and records with a focus on high value cases. The results indicated that, in general, the authorised procedures were followed. However, I have also completed the detailed examination of a sample of cases which I referred to in my Report for 2000. The results of that examination are reported in the following section of the Report.

Mr. Purcell

Sections 4.1 to 4.6 of my report bring together in summary form information on tax collected and arrears of tax, together with a commentary on revenue audit, investigation, prosecution and write-off activities during the year of account. Net tax receipts for 2001 were up €800 million on the previous year while the amount of tax outstanding remained at about €1.5 billion at the end of May 2002. Revenue reckons that it will ultimately collect about 70% of that figure and that is an improvement on previous expected collection performance. Although the number of audits undertaken slightly increased in 2001, the resultant tax levied was reduced because of the distorting effect of the receipt of DIRT arrears in the previous year. I draw the committee's attention to the paragraph on random audits on page 15. These are important as they give us a rough handle on how the self-assessment system is working in terms of the veracity of the returns made. It is interesting to note that in 2001 one out of every three returns randomly selected for audit resulted in additional tax liabilities being assessed. That represents an average of almost €15,000 in the cases where additional liabilities were established.

Section 4.4 records that Revenue successfully prosecuted four cases of serious tax evasion in 2001. At the end of that year there were 31 cases on hand. My information is that three cases were decided in the courts in 2002. A suspended sentence was imposed in one of these cases and fines were imposed in the other two. I will make an observation. Even allowing for the volume of work that goes into mounting a successful prosecution, on the face of it the fact that only seven cases of serious tax evasion have been finalised in court in the past two years is unlikely to be sending shivers of fear down the spines of those engaged in significant tax evasion.

Section 4.5 gives information on the status of the special investigations being carried out by Revenue. In 2001, Revenue began the process of gathering the underlying taxes owed by bogus non-resident deposit account holders. Some 3,675 account holders availed of the voluntary disclosure scheme that closed in November 2001 and paid more than €227 million. Revenue is now pursuing the remaining bogus non-resident account holders. I understand that a further amount of approximately €160 has been received to date. Work is also continuing on the other special investigations listed.

In section 4.6 members will see a further €140 million was written off in tax arrears in 2001. This represents an increase on the previous year which is mainly due to the automatic clear out of a large number of low value arrears as part of Revenue's strategy of concentrating its resources on the collection of current taxes and collectable arrears. The 33,791 cases written off automatically form part of the total of the 35,173 cases which it was uneconomic to pursue.

I ask Mr. Daly to make a brief statement.

Thank you, Chairman. I appreciate the opportunity of making this short opening statement. I would also like to thank the Comptroller and Auditor General and his staff for the thoroughness and professionalism of their audit. I hope our reaction to it has been positive and we will most certainly take on board - and, indeed, have already done so on a number of issues - any recommendations that will assist in improving the effectiveness and efficiency of the tax collection system.

I will deal with paragraphs 4.1 to 4.6 initially. They cover areas which are subject to regular annual audit by the Comptroller and Auditor General and the remaining three paragraphs on the agenda today are project audits. With your permission, Chairman, I will now comment very briefly on paragraphs 4.1 to 4.6 and I will make my comments on the remaining paragraphs as they arise.

Thank you, Mr. Daly. That is acceptable.

Paragraph 4.1 shows net receipts for 2001 at €28,015 million, an increase of more than €800 million, or 3%, on 2000. While this amount is substantial, the rate of increase is less than in recent years reflecting the slowdown in the economy generally.

Paragraph 4.2 deals with outstanding taxes and PRSI and shows the outstanding tax as at 31 May 2001 to be €1.508 billion which is a marginal increase of €15 million, less than 1%, in the nominal value of the debt over 2000. However, given that tax revenues increased by 3% in the period in question and that the change to the calendar year resulted in a once-off adjustment in the debt figure, this marginal increase of less than 1% means that we are continuing to make good progress in reducing the outstanding tax in real terms.

More importantly, we now estimate that we will collect 71% of the outstanding debt - up from 63% a year earlier. It also means that debt expressed as a percentage of gross collection has again fallen from 4.25% last year to 4.01% in 2001. Just a decade ago this figure stood at 40%. We are fully committed to continuing to reduce the debt level and have the following very specific targets in mind for that: to have all debt on record less than six years old or being the subject of active enforcement or court proceedings; and to have not more than 25% of the debt on record greater than three years old.

Paragraph 4.3 reports on the Revenue audit programme. While the number of audits is only marginally up on the previous year, the tax, penalties and interest are up substantially by some €70 million or 50% when the yield from the DIRT look-back audits of financial institutions in 2000 is excluded. In addition to the €208.5 million audit yield mentioned in the report, Revenue auditors collected over €76 million in arrears for periods other than the periods audited. The programme during 2001 therefore yielded €284.5 million.

Paragraph 4.4 deals with Revenue prosecution activity. Four cases were decided before the courts in 2001 and convictions were obtained in all cases. A prison sentence was imposed in each case, two of which were suspended. In addition to this there is, as detailed in the report, an ongoing major campaign of prosecuting persons who fail to file their tax returns which has resulted in over 1,100 convictions and 476 P35 non-filer cases where civil penalties were recovered through the courts. On the customs and excise side there were 400 convictions for various offences, including six cases where jail sentences were imposed. In all, fines totalling €1,970,000 were imposed as a consequence of these convictions across all areas of Revenue activity.

An event of some significance in our approach to prosecutions was the establishment, under our organisation restructuring, of a dedicated investigations and prosecutions division during 2002. Besides increasing the focus on prosecutions, this development will bring together the work of taxes and customs prosecutions into a single division. We now have a clear policy of taking more prosecutions in cases of serious tax evasion and we are getting good results, although I note and accept the point made by the Comptroller and Auditor General that it is a slow process. This is important both in regard to the penalty aspect for offenders and the exemplary aspect for other non-complying taxpayers. I must mention however - and again I go back to the point made by the Comptroller and Auditor General - that this is a very difficult area where it is necessary to build up skills and experience and where the evidential burden is extremely onerous and resource intensive.

Paragraph 4.5 deals with special investigations. I have already submitted a report to the committee which deals, inter alia, with the ongoing bogus non-resident accounts. The big three investigations, namely BNRs, offshore investments via NIB and Ansbacher have already yielded tax including payments on account, interest and penalties of €673 million. In addition we are of course monitoring developments at the Flood and Moriarty tribunals and have a considerable number of cases under inquiry as a consequence.

All of these investigations are resource intensive and some of them are very complex, but this will not deter us. We will see all these through to a successful conclusion. I regard such an outcome, where Revenue is seen to persevere and not to walk away, as vital not just for the credibility of Revenue, but also for changing attitudes to tax compliance in Ireland for the better.

As regards paragraph 4.6 which deals with write offs, or what perhaps in the general business environment might be classified as bad debts, I have already given a brief comment to the committee at the meeting of 13 February and, with your indulgence, I will just reiterate the comment I made on that occasion. Revenue, in common with all other tax administrations world-wide, most large financial institutions and indeed many commercial firms, has to suffer a certain level of bad debt. Not all business ventures are successful - some will fail, people get into financial difficulty. It is a fact of commercial life that certain bad debts will have to be written off. While we would prefer the figure to be nil, I believe that the €140 million written off in 2001 should be viewed in the context of gross tax collection of nearly €38 billion that year. Therefore, tax written off represented 0.37% of tax collected. The most important issue for Revenue is that we tackle this problem at source by taking timely and robust action against late payment and non-payment of tax. Where we have to write off tax, we will do so only when our most intensive efforts to recover it have been exhausted.

Thank you, Mr. Daly.

I join in welcoming Mr. Daly to the resumed examination. On revenue collected, I note that, from the proceeds of tobacco excise duty, €168 million was ring-fenced for the Department of Health and Children. How is the 1999 Act initiative working out? While the full figures for last year may not be available, isthere an increase or a reduction and how is it being accepted generally? Is it seen as a good initiative?

In terms of whether it is a good initiative, the Deputy is taking me into a subjective area as to its effectiveness as a deterrent.

Is it working fairly effectively?

From our point of view, it is working, as a levy, in terms of procedures, collection and processes. While I have not got the figures, I believe the amount has been increasing. As to whether it is having a deterrent effect, as intended, I am not sure. I should say it is a fixed amount of €168 million each year. That is what the legislation states. As far as I know, it goes towards funding the tobacco control office. The question as to whether it is working is, I suppose, a matter of how well that office is working and what effect it is having on smoking. Generally speaking, the trend in excise receipts from tobacco has not been downward up to now, although there is some evidence in recent months that that may be changing.

My question was prompted by the impression within Revenue and the Department of Finance that such an initiative might set off a tremor, so to speak. I wondered if it had proved successful. However, I will not pursue the point.

In relation to outstanding taxes and levies, the situation is quite promising, as Mr. Daly has indicated. The expected level of recovery has risen to 71%, compared to 63% in May. What factors have caused such a substantial increase over a period of 12 months in the expected yield?

It is really a feature of a more aggressive approach to late payment and non payment of tax. Formerly, we had a serious debt problem in Revenue, where the debt was about 40% of gross yield each year. That is now down to about 4%. In recent years, we have had a game plan which is very straightforward. First, we try to eliminate the older, entrenched debt, of which there was quite a great deal in the system. At the same time, we have to avoid the accumulation of any new debt. Eliminating the old debt simply involves a more aggressive approach, based on a few projects we have initiated in recent years to "get stuck into" the older debt, either collecting it if it is collectable, discharging it if there were estimates which were not tenable in the first instance or writing it off if there is really no prospect of collecting it. That is in relation to the older debt.

To stop new debt accumulating, we intervene earlier. That is the solution to it - we do not allow people to accumulate debt any more. We monitor the performance of companies, businesses and individuals who owe us money. In the Collector General's office in Limerick, we have a system of early pursuit and we have local collection officers throughout the country who will go knocking on doors if debt is accumulating. We have been much more aggressive in the use of sheriffs, solicitor enforcement and what I might call the use of exemplary initiatives on the debt front. That would include, in some cases, being more active in the liquidation of companies. In some cases, not very many, we have considered the idea of bankruptcy and we also use attachment and court instalment orders. I can provide the relevant figures. In 2002, we sent nearly 8,000 cases for solicitor enforcement and nearly 30,000 cases for sheriff enforcement. We have used attachment in 689 cases, we have obtained nine instalment orders from the courts, we have undertaken 12 forced sales of property and we have had nine bankruptcy proceedings.

It is really a combination of all those approaches. A comment was recently made to me to the effect that we are very much on the ball now, not letting people away with anything and writing to people who are only a month in arrears. However, we are still not resting on our laurels, at just over 4%. Despite the increase in gross collection, the increased level of activity in the office and the increased volume of business, our aim, for the year 2002, is that the debt figure will be further substantially reduced, perhaps by another €200 million to €300 million, to well below the 4% level. I would be very happy if, within one year, we could bring it down to 3% but it may be a little early to say at this stage. We will have those figures in about a month's time. However, the indications are good.

Over the years, there has been some resistance to writing off bad debt. In relation to the 33,791 automatic write-offs, does that represent a change of mind-set? It is a huge number in what I presume is a once-off effort. Is it a changed approach?

It was, perhaps, a pragmatic approach rather than anything else, following discussions with the office of the Comptroller and Auditor General. We had a great number of relatively small amounts, mostly quite old. To actually case-work each of them would have tied us up in knots for years, thereby restricting our capacity to tackle the really big debts. Accordingly, with the understanding of the Comptroller and Auditor General, we developed a programme which automated the write-off process for certain years, subject to upper limits and spot checks on individual cases. It was, to a very large extent, a one-off project but it is not quite finished yet - we still have some work to do. If we get to the stage where we have the debt within manageable proportions, I do not see automated write-off as being a regular feature. I prefer to get to the stage where it is manageable and where we are intervening in cases, with an actual "eye-balling" of every case in future.

I agree there is a more pragmatic approach. The emphasis should probably be on paragraph 4.3, i.e. the audit programme. There is a very worrying trend in the 2,200 cases that were the subject of a comprehensive audit, with only 813 being found to be correct. In 409 cases, the amount owed was between €12,500 and €63,000, in 93 cases, it was between €63,000 and €126,000 and, in 105 cases, the amount was more than €126,000. That represents a huge proportion of cases. There were only 577 cases where no additional tax was payable and others came into the net for various reasons, leaving a total of only 813 correct. It is worrying on face value. Can Mr. Daly explain why it is so bad?

It may appear bad, taken at face value, but it must be remembered that most of our audits other than the random audits are targeted. We have gone through a screening process and we have tried to choose cases that are most likely to have a yield - they are the riskier cases. It is probably useful to give the committee an idea of how cases are looked at in the districts each year. The case base for audit is quite extensive - there is a base of about 400,000 cases for income and corporation taxes that should, ideally, be looked at every year if we could audit everybody. Approximately 2,000 large cases are screened in each district and audited every second year. All CGT cases are given special attention - there is virtually 100% screening. All other cases are screened over a three to four year period, depending on the perceived risk. The audit managers choose the cases for audit from the 70,000 cases that are screened each year. By the time the cases are selected for audit, they have been through a process that has tried to identify the risky cases. In a way, that explains why one will receive such a return in so many cases.

The Comptroller and Auditor General referred earlier to the random audit programme which is in place, in the first instance, to emphasise that anybody can be audited in any year. It also gives an indicator of overall compliance levels to check the effectiveness of targeted audits. That the return from random audits is not as good as the return from targeted audits may answer the Deputy's question. I accept the point that the Revenue Commissioners receive a return in too many random cases. I would prefer if the random cases were indicating that there is a better overall level of compliance. It is something we have to work on.

Investigations under the random audit programme provided a return in 230 of the 472 cases that were examined. I appreciate that one is dealing with a bad pack in the targeted programme, but I am concerned by the finding of our previous discussion - that certain people are not listed anywhere. Are people who are not on the list - who are not in Revenue's net - engaging in trading?

It is difficult for anybody to trade for a reasonable period of time without the knowledge of the Revenue Commissioners. I recall that the Comptroller and Auditor General made a comment to that effect when he produced a report on Revenue's compliance operations. Part of the focus of the special inquiry branch in Dublin and the local inquiry units in all tax districts is to ensure that people cannot trade or do business without coming to the notice of Revenue. These matters will be strengthened in the new structure, which will include special compliance districts in every region. There will be a focus on knowing what is going on and ensuring that nobody can trade or do business without being on Revenue's books. I am fairly satisfied that it will be difficult for people to continue to evade the attention of Revenue for any length of time. I know we cannot stop people from trying and they may be able to get away with it for a few months.

We can send out the Comptroller and Auditor General with a trawl sometime. I am sure he will pick up a couple of them.

It is inevitable that one will always find somebody. The most important thing is that the Revenue Commissioners eventually get them on their books and then assess the matter retrospectively.

This committee has been interested in DIRT. The Revenue Commissioners have rechecked 47 audits of the financial institutions involved to ensure that they were keeping up with what was not done in the past. How did that work out?

The audit of financial institutions' current compliance with the DIRT regulations is not showing up any difficulties. Compliance with the regulations, as far as we can ascertain from our current audit programme, is very satisfactory. I do not know if the committee wants a more general update on the bogus accounts——

It might not be any harm, in light of the experience with Ansbacher and offshore accounts. I appreciate that the Revenue Commissioners have placed their report before the committee, but I would like Mr. Daly to fill us in about the matters he has mentioned.

The committee will recall the various phases of the investigation into bogus accounts. The first phase, during which we collected €220 million, involved an examination of the institutions, including the banks. The second phase involved collecting the underlying tax from the account holders. We sought High Court orders in respect of all 26 institutions that appeared to have the possibility of bogus non-resident accounts emerging from the DIRT inquiry. All 18 High Court orders that are required in respect of the institutions have been acquired. The first part of our second phase was the incentive scheme, which ceased in November 2001. The committee will be aware that €227 million was paid by 3,675 account holders in that phase. Since then, inquiry letters have been issued in two major tranches, in October 2002 and January 2003. The first one was issued in respect of 13,500 accounts and the second was issued in respect of 21,500 accounts. This means that 35,000 accounts are under inquiry at present. Some €159 million has been collected in this phase of the inquiry to date.

The total yield from the DIRT inquiry is about €606 million - that is this morning's figure, as it were. The Revenue Commissioners receive information from the banks and will continue to do so for a number of months. There will be more inquiries in the coming months and more letters will be sent out. The level of engagement with us from those in relation to whom we make inquiries is quite good, generally speaking. It has transpired that quite a few people do not have a liability or a substantial liability and that clears them out of the system. It has been speculated that this investigation could continue for five or ten years, but it is my ambition to deal with it as quickly as possible. There will always be stragglers in any investigation, but I hope to have the bulk of it sorted out some time next year.

That is very good news.

I appreciate that it is a long haul, but it is important that this committee is kept up to speed. I presume the 35,000 cases that were automatically written off, a more than fivefold increase, do not include those that were uneconomic to pursue. Is that the same pragmatic approach?

It would be uneconomic to pursue them.

The Revenue Commissioners wrote off 35,173 cases.

It is a pragmatic approach, by definition. The amount of money involved in the cases is not worth pursuing, from a value for money or opportunity cost point of view. One must consider what might be done elsewhere and the amount of tax and the effort involved.

Are some of them old cases or do they all relate to 2001?

Some of them could be quite old.

Is that going back to 1991?

I can get the Deputy a profile of that.

Mr. Purcell

The 33,791 cases written off automatically form part of the total of the 35,173 cases which it was uneconomic to pursue. To clarify, they are separate figures.

That was really the question. I am concerned about people who do not return employee's material such as the P35. A very good advertising campaign is being run and fines of over €1 million have been imposed, but are the Revenue Commissioners happy with the level of compliance? A serious aspect of this is not so much the way it relates to Revenue, but in relation to employees. I note that four people were jailed for other offences.

Regarding the compliance on the P35 filing, I would never say I was 100% happy with less than 100% compliance. On the due date for P35, returns run at about 65%. Within a week compliance levels reach 80% and by the end of the month compliance has been of the order of 92%. While that means that 92% of all employers required to return P35s have done so, a much higher percentage of total employments will have been covered in that figure. By and large, those who do not make returns are small employers or those who have ceased to trade.

We invest a great deal of effort in ensuring that P35 compliance is monitored and the process is the subject of our regular annual advertising campaigns. Much activity is carried out from our office in Nenagh which is responsible for the matter. It is one of the areas in which compliance is satisfactory. I may have misled the Deputy by reading from a certain column. Rather than deal with the current year, for the later months of which I have no figures, I should go back to the previous year. Due date compliance in May last year was about 58% which had increased to 90% by the end of the month. By the end of the following March, compliance had increased to 99.6%. It is satisfactory.

I have a question which reflects political activity with regard to the Ansbacher pick-me-up schemes and the 55% of persons found to be culpable. Were they named and shamed in the same way as other defaulters? There seem to be 14 outstanding cases as of 56 cases, 42 have been settled. What is the situation with those?

There were 71 cases in all in the scheme. Of those, 15 have been mentioned at either the Flood or Moriarty tribunal, which is why we have not finalised them. While they will be dealt with, it is inappropriate to bring them to a conclusion at this stage. Of the 56 other cases, 42 have been settled and in 32 of those the tax interest and penalties did not exceed the publication threshold, which is €12,700. In six cases the tax interest and penalties exceeded the publication level and of those, three were published. The other three were not published because tax and interest had been paid at a time when the legislation permitted someone who paid a 100% penalty to avoid publication. The legislation was subsequently changed.

Are there still 14 cases outstanding?

There are 14 cases still under inquiry, some of which are quite difficult as they relate to transactions which took place in the late 1980s. Records, including audit records, are no longer available.

They might end up being written off.

Many hoops will have to be jumped through before they are. It is proving very difficult to confirm liability. In eight of those cases the liability would exceed the publication limit if we could prove it. The cases remain open.

Are we prevented from having circulated the names of those involved in the cases referred to at the tribunals? Can we get a note on them, or are they not in the public arena?

They would have to be in the public arena. Taxpayer confidentiality comes into play in this case.

I thank Mr. Daly for his comments. Whether people agree or not, there is a new pragmatic approach in the Revenue Commissioners to certain issues. I am glad of that.

The gross tax take of €38 billion is described in the report as €32.5 billion. The net is reported as €28 billion.

The gross versus net difference is the result of repayments in between. The difference between the figures of €38 billion and €32 billion is the result of PRSI.

What are the repayments?

A substantial amount are VAT repayments of which there are quite a high number.

What percentage of the refund is VAT?

I do not have the percentage, but €2.6 billion was repaid in VAT and the other large repayment was €1.2 billion in income tax where overpayments were made.

Is that figure €3.8 billion? Is that the total repayment?

There are other bits and pieces such as excise, corporation tax and stamps, but those two make up the bulk. I do not have percentages as such.

I apologise for my late arrival. I was speaking at my daughter's school in Cork at 9.10 a.m. My journey has been of "Star Trek" proportions.

The Deputy must be a fast driver.

I welcome Mr. Daly and his team back to the committee. My question relates to the audit programme and the variations contained in the figures provided to us. The number of audits has increased between 2000 to 2001 by 15, yet the number of targeted audits in specific tax areas has decreased markedly. There have been decreased audits of VAT while audits of PAYE and income tax have decreased by over 30%. Decreased numbers have also been noted regarding contract tax, capital acquisition tax and in combined audits of all these taxes. Audits have increased, however, in verification audits and desk reviews which appears to be a surface scratching exercise which requires less resources and time to conduct. Can Mr. Daly explain the variations in the two sets of figures?

We are looking at table 4.

That is right.

During 2001, we began to deal with legacy investigations of bogus non-resident accounts, the Ansbacher pick-me-up schemes and the NIB schemes. We were also dealing with the DIRT inquiry and the tribunals which took, and continue to take, much of our experienced audit resources. It was a large overhead for us at the time.

At that stage, the Minister also approved an increase in the number of staff, most of whom were allocated to auditing. However, the new personnel had to be trained by experienced auditors. The results of this increase in staff will probably emerge in the coming years.

There has been a slight fall of 70 in the number of comprehensive audits, by far the most onerous and difficult of audits given that some of them take considerable time to complete. One would need to examine the composition of the 2,200 completed audits as just one of these could take several months to complete.

The same probably applies to the area of value added tax. It should also be borne in mind that certain matters placed a major drain on Revenue resources in 2001. We were also putting in place perhaps a better cadre of auditors. I have figures on the number of staff involved in audit. During 2000, for example, some 746 staff were involved. In 2001, the figure was ratcheted up to 824 as a result of the extra staff approved by the Minister and in 2002 the figure increased to 876.

Will Mr. Daly repeat the figures?

The figures for staff working in audit in 2000, 2001 and 2002 were 746, 824 and 876, respectively. It takes time to become an experienced auditor and the process intrudes on the time of experienced auditors who have to mentor and train new staff. I have never been hung up on the numbers of audits as audit is just one tool in Revenue's compliance agenda. While it is a useful comparator of activity from year to year, I am much more concerned by the impact the audit programme is having on compliance.

In this context, the Deputy has a point on verification audits and desk reviews, which I am not certain always represent the most effective use of auditors' time. The report contains a project audit on VAT repayments, which are often the subject of such audits. Revenue must at all times make calls on what is the best use of resources. Perhaps we should not undertake so many verification audits and instead redeploy staff to comprehensive audits. One explanation for the number of such audits undertaken in 2001 is that staff generally begin training in verification and similar audits before moving up to comprehensive and fiduciary audits.

Perhaps next year's report will show a different approach. Would it be fair to assume as a rule of thumb that average salary and administration costs per auditor are in the region of €100,000?

Various grades and levels apply to staff.

If we take this figure as a rule of thumb, it appears that in 2001 the administration of audits cost some €82 million and €208 million was collected, whereas in 2000, €75 million was spent on administration and €359 million was collected. The value obtained in terms of the moneys collected versus administration costs appears to be declining. Has this trend been noted and are steps being taken to correct it?

The audit programme is designed to do more than just collect large sums, such as the €208 million the Deputy mentioned, each year. It is designed to act as a deterrent and make people more compliant. It feeds into the area of tax collection. One must also factor into the figures the additional significant resources allocated by Revenue to other investigations initiated since 2000-01. The investigation into bogus non-resident accounts is probably one of the largest investigations undertaken by any tax administration in such a short period and we have not yet discussed the Ansbacher and NIB investigations. These were all complex investigations which required considerable investment in terms of time and effort. For example, Revenue had to seek High Court orders. Much more was required than visiting the companies in question and going through their books. A major back-up operation was also necessary.

If one was to carry out a cost-benefit analysis of the cost of auditors, one would have to consider not only the €200 million or €300 million directly collected, but also the arrears, payments on account and, most importantly, the impact of their work on general compliance. Does it, for example, lead to people paying more tax or becoming more likely to declare correct tax liabilities? The general indications are that this is the case.

The fact that €150 million less was collected in 2001 than in the previous year could be indicative of a trend of general tax compliance throughout the system. However, surely other factors related to greater tax buoyancy must also be considered. More money is being collected and paid. A drop of €150 million in one year indicates that some other factor is at work. Did it send out warning signals?

The Deputy will note that the DIRT investigation yielded €220 million in the year in question. To make a true comparison, one would have to——

I was looking at the headline figure.

——exclude this figure. This is the reason I made this point in my opening statement.

My mind is somewhere else. The Minister for Finance gave new powers to the Revenue Commissioners in successive Finance Acts. Does Revenue monitor the extent to which each of these powers has been used in recent years?

Monitoring is undertaken in certain cases. While I do not have the figures to hand, I can supply them to the Deputy. We have a range of powers, some of which are only exercised in extreme cases and are subject to approval at a relatively high level in the organisation. An undertaking to this effect was given in 1999 at the time the Minister introduced the relevant powers and is a necessary control to ensure they are not abused in the Revenue Commissioners. We have figures on the use of these powers. There are also routine powers used on a daily basis by tax inspectors and customs officers. It is unlikely we would have details on the use of such powers.

As the committee will be aware, in recent weeks the Minister established a review group to review the powers available to Revenue to establish tax liabilities. The group is headed by Judge Frank Murphy and must report back to the Minister by October. When we were given these powers in 1992, it took a while to get a feel for them and to identify how they could be best and must effectively used. It was also necessary to balance their use against citizens' rights. We established procedures to ensure the higher level powers were used carefully and were monitored. The experience we have gathered in the intervening years places us in a position to review the level of approval required to exercise such powers. I would like to encourage greater use of such powers by tax inspectors. This review process is under way. It is likely these powers will be more widely used in future and will become largely routine in the various tax districts.

Would it be possible to supply figures to the committee on the use of current powers to give us a greater understanding of the way in which they are used?

I can give the Deputy some figures now, but may have to get back to him with others. In 2001 our prosecution units obtained 26 orders for disclosure by financial institutions under section 908 of the Taxes (Consolidation) Act and three search warrants for the purpose of investigation and collection of evidence. There were six orders for disclosure under section 908 and 14 search warrants in 2002. These were by our prosecution unit and probably do not give the full picture of the use of powers right across the organisation. Therefore, it might be better if I wrote to the committee on the matter.

I appreciate that. Is Mr. Daly aware of an internal difficulty in the Revenue Commissioners, possibly at the level of the inspector of taxes, concerning a lack of individual decision making power because of the need to refer upwards decisions that could be made on the ground? Leaving decisions to the office of the chief inspector of taxes sometimes causes delays.

In recent years, I suppose I have heard of such cases but we have gone through a process of restructuring in the Revenue Commissioners over the past couple of years, which has been accelerated since 18 March with a very significant step forward. One of the main thrusts behind that reorganisation and restructuring has been to devolve decision making from the centre to five regions.

I am not sure that there was a huge barrier to decision making in the old structure. Certain things had to go to the top and maybe that was an attempt to ensure consistency in the application of the tax code, which must be consistent throughout the country.

We have reviewed all our structures and we are in the middle of reorganisation and devolution. The devolution of decision making from the centre to the regions represents a very definite policy decision by the Revenue Commissioners. If there are concerns, this will more than take care of them but I am not absolutely convinced that there was a problem.

I apologise for my late arrival. Consider table 15 pertaining to collection and enforcement activity in 2001 by solicitors, sheriffs and attachment orders on page 41 of the annual report——

We are dealing with paragraphs 4.1 to 4.6 of the report.

My apologies. I will raise my questions later. My other query concerned dividend withholding tax. Is that covered under the paragraphs with which we are dealing?

That will arise later.

Mr. Daly mentioned a round figure of €673 million accruing from three special investigations in respect of taxes owed and penalties pertaining to DIRT, National Irish Bank and Ansbacher. Does he have a breakdown of what has accrued from each individually?

I have. The bogus accounts figure is €606 million. As of today, the yield for the National Irish Bank CMI scheme is €44.5 million, the Ansbacher yield is €21.86 million and the total yield in respect of the pick-me-up schemes is €720,000. If one adds those figures together, one will get approximately €673 million.

The Ansbacher issue was debated a number of times in the Dáil and was obviously the subject of considerable public interest and amazement. It involved a very powerful elite. A substantial multinational corporation was virtually running an illegal bank, set up deliberately to defraud the tax system. Have there been any prosecutions of those involved in the Ansbacher adventure? Unfortunate constituents of mine in the District Court get hauled over for shoplifting, which of course they should not do, and get jail sentences. Where Ansbacher accounts are concerned we are dealing with a very powerful and wealthy elite and considerable sums of money. What is the current position?

The current position on Ansbacher accounts is that we have 289 cases and there are 700 entities connected to those. There are currently 211 active investigations and there have been six settlements, which is not very many. Some 62 of the 289 cases are non-resident, 12 are amnesty cases who availed of the 1993 amnesty, and there are four cases in which we have not been able to establish identity sufficiently.

We have had a huge investigation since 1998. We have gone to the High Court, we have got five High Court orders and we have used other powers extensively. We have 150,000 documents as a result of the orders and impending High Court application for the inspector's papers. The application was heard on 26, 27 and 28 November 2002 but judgment has been reserved so we do not know what the outcome will be.

As I said, the yield from the Ansbacher investigation to date has amounted to about €22 million. It is proving to be a very complex and intransigent investigation but we are pursuing it and will not let it go. The Deputy asked specifically about prosecution. If we can, we will prosecute people who were involved in the Ansbacher scheme. However, one of our main problems is that much of the evidence and documentation is offshore and some of the people we need to bring to court are either offshore or, in some cases, dead. While the Deputy can take it that the Revenue Commissioners do not lack the will to prosecute if they can, they must be realistic about their chances of doing so.

I do not doubt the deviousness or resources of the people in question. The scheme was, by common consent, ingenious and devious. However, is it intended to pursue the substantial estates of those who have shuffled off their mortal coils, as it were? How many of those who reside offshore still have citizenship of this State? What is the legal position when the Revenue Commissioners try to get such people to answer questions they need answered?

We will pursue their estates in the normal way. If anyone has a tax liability, and it is established by the Revenue Commissioners, we will go after the estate and its beneficiaries. I do not have information about the Deputy's question on citizenship. All I know is that we are not leaving aside the 62 people that are claiming non-residence. It is not enough for them to tell us they were non-resident, we will inquire into it to verify such claims. There could have been people who were, in tax terms, genuinely non-resident.

I may have inadvertently given an overly pessimistic view. The €22 million has been paid in respect of 69 cases, of which only six are finally settled. It does mean that 63 others have at least engaged with us. In most investigations we can expect a certain degree of co-operation and a degree of realism will creep in. In this investigation we are being fought every step of the way by many of the people involved. It does not deter us and we will keep going.

Does the citizenship of this State, or otherwise, of the 62 non-resident entities have any affect on the ability of the Revenue Commissioners to gain access to them? If such individuals are citizens and are not co-operating, what is the position when they arrive back here for special occasions etc.?

I have to be careful about saying anything that might inadvertently point to anybody. As far as tax matters are concerned, residency, not citizenship, is the primary consideration. When I say there are 62 people claiming non-residence, from looking at the Ansbacher report, it is not clear to me that we might have doubts about the residence of many of those. Where someone claims non-residency, non-liability or anything else, full inquiries are made. If someone claims they were non-resident in Ireland for tax purposes - this is a better approach than the citizenship one - we do not accept their word for this. We will inquire into this and seek to establish their tax status in the country where they claim tax residency.

Do these people have to be pursued through the laws in the jurisdiction in which they reside, rather than here? Is there an estimate of how much is outstanding from those who have not yet had a final reckoning?

We do not have an estimate as in many cases it involves wading through documentation and one cannot estimate tax until one is sure of what the liability or founding income was and as to whether that income was taxable in the first place and if someone was genuinely non-resident for tax purposes.

The first port of call for us is our powers in this jurisdiction. As regards foreign jurisdictions, we are getting to the stage where there will be no hiding place for people who want to hide money offshore. We are getting to this stage because of developments within the EU. Under the savings tax directive, and with effect from January 2005, most member states will routinely exchange information about savings by residents of other member states to those member states. In the three member states that are not subscribing to this, a withholding tax will be applied in that jurisdiction and Ireland will receive 75% of it.

The OECD has been doing sterling work, aided and encouraged by the Revenue Commissioners, in trying to eliminate the concept of tax havens. By means of tax information exchange agreements, most of these tax havens are now in the process of becoming the equivalent of good citizens, if I can put it that way. There are still four of five places that are out of the fold. In the next year or so, the Revenue Commissioners will be able to exchange information from places that would traditionally have been known as tax havens. We have already initiated discussions with the authorities in the Cayman Islands, Jersey, Guernsey and the Isle of Man. There is no place to hide and it is becoming increasingly easier for us to access information about people that have moneys hidden in other jurisdictions.

On page 13 of the report of the Comptroller and Auditor General, the figure of €659 million is mentioned. How collectable is that money given that is has been outstanding for so long? Some of it dates back to 1990.

It is over ten years old. I am not sure that I have an estimate of what proportion of that is collectable. I would not see all of it as being a lost cause. Much of it is being aggressively pursued right now.

In your opinion, what percentage would be obtainable?

We are now saying that we hope to collect 71% of 72% of all outstanding debts. While this figure can be applied generally, it decreases the further one goes back.

Would many of those companies still be trading or would some of them be limited companies that have ceased trading?

While I am not sure, I would imagine that some of them might still be trading.

At 31 December 2001, income tax was down at €2 million.

It must be an entrenched €2 million that we have not been able to get our hands on. If you were to ask me about a recent figure of €2 million, I would say that it is probably collectable.

Why is the figure so low?

It might be something to do with post-amnesty.

Mr. Purcell

It may be that since the nine month period to the end of 2001 is covered and the date of this figure, May 2002, is only five months later, a lot of it would not have fallen into the debt category at that stage. I imagine that it why it is so low.

It was the short year.

Has the integrated taxation processing system been extended to all taxes yet?

By and large, yes. The last time I was here, there was a question mark over whether it had been extended to relevant contracts tax. That has certainly been done. It has also been extended to capital gains tax and it has been and continues to be a great success story for us. Since I was here last, it has also been extended to the customs and excise side, including vehicle registration tax. That came into effect this year and was available from 1 January for new car registrations. The take-up has been considerable.

With regard to the Revenue on-line service, we have something like 16,000 registered customers, many of them tax practitioners in accountancy firms. The most important thing is that they represent about 70% of the relevant taxpayer base. That is quite a good take-up. We have collected just under €7 billion electronically since ROS was set up and have made electronic repayments of €578 million. We have received more than 200,000 returns through the system and the total number of transactions through the system is approaching 850,000. It is probably one of the real success stories of the Office of the Revenue Commissioners in the last couple of years.

Why did the report refer to the office's policy of rotating staff amongst customers in 2001?

Does this refer to VAT repayments?

It emerged in the Comptroller's report on the VAT audit that the same staff were dealing with the same tranche of customers in the VAT repayments area and that this was not perhaps the most healthy practice. We have taken that on board and the rotation process has already started. There will now be regular rotation every two or three years.

On page 15 the report refers to the random audit. What is the unit cost of this? The aim was to do 6% but the office ended up doing many hundreds.

We did more than the target number. I do not have a unit cost but I can try to obtain that for the Deputy.

What is physically involved? Are those random audits extensive?

They can be extensive. They cover the different types of audits: comprehensive, fiduciary, VAT and income tax. The same type of audit is done on a random basis as on a targeted basis. There is the same notification to the taxpayer, the same visit by the inspector to the taxpayer's premises, and depending on the size and complexity of the case, the audit can take as much or as little time as any of the targeted audits.

Does Mr. Daly have no idea of the average - given the different sizes of companies - unit cost of doing one of those?

I do not. As I said to Deputy Boyle, an audit could go on in a company for three or six months while another could be finished in a day.

That is why I asked for the average cost. I do appreciate the point Mr. Daly is making.

We can try to come up with average figures, but I would caution against them being used as the be all and end all——

I do not intend to, but I do think it is an important figure because we know the average return obtained by the office. It is a measurement figure. Is the exercise in any way productive or should it be extended? We know there was one return, which was about €15,000, amongst those that were found to have a liability, but we do not know what the cost of that was. We should consider this, because other questions arise from it, such as whether the 6% policy should be extended based on the type of return. For this reason I would be interested in the average rather than an individual figure. I appreciate that some audits take one day and others take three months.

It is a fair point. Mr. Clayton has just reminded me that we are fairly advanced in the development of a computerised risk analysis system, which will drive the whole selection process for audit. Part of that system will be quite a lot of in-built management information such as the Deputy is talking about, which we do not have now, so that will be much more at our fingertips in future. However, I will not wait for that but I will try and work out some figures.

That could subsequently have an impact on how the random audit process is driven.

The findings from the random audit were that roughly one in three organisations had a liability the value of which we know. How, then, is that information used in relation to the some 90% that the office does not audit? I presume it would deduce that amongst that 90%, one in three would have a liability. How are these tackled?

The main way in which we do so is that the results of the random audit, which will throw up a pattern in a certain sector, business, industry or section of taxpayers, are fed into the profiling or screening system. The general results emerging from the random programme inform the next year's selection process.

Would I be right in saying, based on the random audit figures, that 90% have not been audited and of that, 30% would have a liability if an audit was done?

I do not know whether it would be valid to make that assumption.

On the face of it, it seems correct, and this is something that is of concern to us. However, it must be considered in terms of the entire audit programme. I mentioned earlier that every three or four years, all cases are screened, so that out of that process come the most likely cases from which we could get a yield. They come into the targeted audit process. Nobody likes a situation in which random audits throw up a yield, because it is indicative of a problem in the system. This is why, because of the numbers of cases compared to our resources, the reality is that we will never be able to audit everybody. We have a rolling audit programme, which does mean that at least every three or four years every case is considered, and we have this new computerised risk analysis system which will come up with a much more sophisticated screening process.

Does Mr. Daly still have concerns that despite all of this, based on the random audit, one in three still has a liability?

It is not something I like, but it is a liability, albeit a much smaller one than we get from the targeted audit. I take the Deputy's point, however, and the objective is to improve this as far as we can.

It is a liability, and if the random checks were to be expanded over the entire operation, we could be looking at about €30 million. I take Mr. Daly's points, but I was surprised that with all the other mechanisms in place, a random sample was still coming up with one in three having a liability.

I do not argue with that but the programme is designed in terms of the way we screen, our rolling audit programme and the development of a new and more sophisticated selection process to get to the stage where random audits do not yield anything like that figure.

Mr. Daly will be appearing before the committee in years to come. Is he suggesting that the ratio will diminish as the years pass?

I am wary of putting my head on the block and stating that it will be better next year but that is the goal for the organisation.

We do not know the cost of what we are recovering but I am concerned about a rate of one in three. It is not satisfactory. Compliance procedures should have been tightened up so that percentage would not be so high. If I have interpreted the information correctly, I am concerned.

I hope to return next year with improved figures. We are going through a significant period of change at the moment. Everything we are doing, be it restructuring, establishing a new audit system or a different, more aggressive attitude to prosecutions, is designed to improve taxpayer compliance. That is why I said that I do not get hung up on the numbers of audits compared to their results.

We are going through a process that will dramatically improve compliance in this State. There is a plethora of issues - the audit programme, the compliance programme, more aggressive debt management and the fact that we follow through on investigations such as those into bogus accounts, Ansbacher and NIB. Revenue is no longer seen as an organisation that walks away from anything or that will get tired of looking at someone if that person holds off long enough. We are not in that business.

I agree. My concern stems from the question asked by the Chairman about looking at the older debts. The sooner the debts and liability to Revenue are identified, the better the chance of recovering that debt and compliance. It is very difficult when something is ten years old and the entity no longer exists.

There is a human face to public accounts in terms of the bogus non-resident accounts. Many of us have been approached by elderly people who are getting letters from the Revenue Commissioners about bogus non-resident accounts. It causes a great deal of stress. The point consistently made to us is that these people never knew what an offshore account was or dreamt up a fictitious address and would never have invested in such a thing. It is causing a great deal of strain, hardship and worry to elderly people.

I have no problem with the pursuit of those who were deliberately involved. In the tracking and dealings with people, is it obvious that the banks were complicit in advising these people and that they were directing people to offshore accounts as a matter of policy?

I have no comment to make on the banks. I have no more evidence than anyone else about the banks' role in this apart from what was extensively aired at the Public Accounts Committee investigation into non-payment of DIRT.

I am asking about the field officers who are dealing with these people. What sort of feedback is there? Is it not apparent that many of these people would never know what an offshore account is, not to mention a fictitious address, and that the banks were complicit in directing them in this way?

Our field officers get feedback from all sorts of people about the bogus non-resident accounts. No more than people complaining to politicians or the media, people will tell our inspectors that they knew nothing about this and had nothing to do with it. I do not want to use that to form an opinion about the complicity of the banks. That was aired extensively here some years back.

In general, and particularly in relation to elderly people, we have a job to do that we were endorsed to do by this committee. We were asked to do it in a pragmatic and practical way, to use the committee's own phraseology. Our attempt to be pragmatic was the incentive scheme which finished in November 2001. We advertised that heavily and encouraged people to make use of it through television and radio campaigns, spending almost €1 million on an advertising campaign encouraging people to avail of the incentive scheme. There was no one in the State who was unaware of it, we went to enormous lengths to publicise it.

Accounts were cleared by 3,600 people and €220 million was collected. We made it clear at the time that we would have to follow up on people who did not avail of this. That is what we are doing. I know there are people out there who were less than fully aware of what was going on and that there are elderly people who are concerned. I get letters as well from them and I have sympathy for the circumstances of many of the people.

I mentioned earlier that in many cases the liability is turning out to be nil or very little. From my own soundings, there are many elderly people who have the letter from the bank, then the letter from the Revenue Commissioners and they are sitting at home and multiplying figures by five or ten. It would be much more productive for them to come into their local tax inspectors. We have never put an elderly person out of his or her home and we never will. They should come in and talk to the local tax inspector and we will look at the facts of the case. I sympathise with such people.

I accept there are people who were probably not fully aware of what was going on but we made huge efforts to give people the opportunity to sort this out in November 2001 and we now have no alternative but to keep going. People should not sit at home building up problems or worrying about it, they should come and talk. While we have to collect the money, we are not going to turn people out into the street.

In my opinion, the banks encouraged this business to a significant extent. Many elderly people visited my office and, while I accept that the Revenue Commissioners have been fair, I feel the banks are culpable to a degree. They felt that what they were doing was correct because they had the approval of the bank which initially encouraged them to open the account. Clearly the banks promoted the scheme, encouraged people to open these accounts and carried all the paperwork. I have no difficulty in blaming banks for promoting this scheme when they knew it was illegal to do so. A number of people who came to my office indicated that they were given an address by the bank to open accounts. The banks got off quite lightly, given that they paid €220 million while more than €400 million was paid in tax by people who had a tax liability.

Mr. Purcell

If we are finished on this section, at the risk of sounding as if I have a heart of stone compared to Deputy O'Keeffe's compassionate face of the committee, let me say that what is being sought by Revenue now is the tax on the underlying income in those accounts. Whether or not that money was in a bogus non-resident account or, let us say, under the mattress at home is irrelevant because in a sense the underlying tax is tax that should have been collected anyway. The fact that it came to light through the bogus non-resident accounts is a bonus. That is how I understand it.

Let me go back to a point I originally meant to make, which supplements what Deputy Curran said. I was very glad to hear what he said about random audits. The current Accounting Officer will know, as his predecessors would have known, that this has been a particular hobby horse of mine for many years. While I accept one does not get the same return from random audits as from targeted audits - and targeted audits are very important - it is absolutely essential that a certain quantum of random audits is carried out. It is to the Revenue Commissioners' credit that they exceeded the target in 2001 because there were other years when they did not come near meeting the target. They are important because, while it might not be absolutely statistically sound to extrapolate in the way Deputy Curran has, given the current number of random audits, about 740, if the conditions under which they are selected randomly are sound, one can get close to extrapolating, in the way the Deputy has, that it is likely that the population of those that are not audited is likely to have one in three who are understating their liability in their returns. It is useful to take it that little bit further. By setting up in graphs, computer models and so on it is possible to set down the quantum of inspection or audit resource one needs to apply to that area. I have no doubt that ultimately over the coming years, as Revenue becomes very much more high-tech, that is something that will be examined and will tie in with the question of the cost of carrying out those audits. I would put it further in terms of analysing the optimum level of audit that should be applied to those returns in order to get to a position where it clearly is not advantageous to understate one's liability. I am sorry to have gone on a little, but it is a very important area for the future.

Thank you, Mr. Purcell. Are there any further questions on 4.1 to 4.6? If not, we will move on to 4.8.

Under what chapter does the section on enforcement and collection come in?

Does the Deputy want to put a question on that?

I did earlier, but you told me you thought it came under 4.1 to 4.6.

That deals with the withholding tax.

I had a question on collection and enforcement as well. Is that under this?

Can I deal with that?

You can.

I refer to page 41 of Mr. Daly's annual report which deals with table 15, Collection and Enforcement Activities 2001. It deals with collection by solicitor, sheriff and attachment orders. I would make the following observations, having analysed the chart on page 41. The number of collections by solicitors is 6,106 with an average yield of €8,370 per enforcement. The number of collections by the sheriff was 23,558 with an average yield of €4,280 per enforcement. The number of collections by way of attachment order was 689 with an average yield of €6,488. The least efficient method is the sheriff. Why is there such an inordinate number of collections by this method when collections by way of attachment orders and solicitors are far more efficient and yield far more money? Although we are obviously dealing with taxpayers who have to be dealt with in this manner in the first instance - we are not talking about the vast majority of taxpayers - attachment orders or paper transactions dealing with financial institutions and people who owe money seem slightly less painful and far less traumatic than the sheriff arriving with four men and a couple of vans. It seems a more modern, efficient and humane way to collect money. I would like to hear observations on that and, given that the other methods are more efficient for collecting money should Revenue not rely more on those methods and less on the sheriff?

In essence the sheriffs get the more intransigent debts to collect. In most cases we would have exhausted a lot of avenues before sending cases to the sheriff. That is why the return per sheriff appears not to be as good as that from solicitor and attachment. I agree that attachment is a very productive area, but it is not suitable in all cases. For attachment it is necessary to identify a source of income for the person from whom one is trying to collect a debt. Solicitors, particularly in the past few years when we have engaged new solicitors, have been very effective. However, sheriffs get the most difficult debts to collect. That is the reason they might appear, on the face of it, not to be as productive. I would also make the point in relation to the figure for the sheriff collection that once somebody gets a letter or contact from the sheriff, very often the first thing they do is pay the tax directly to Revenue. I am not sure this is reflected as a return through the sheriff. It may be that this is misleading. I am not 100% sure of that but I can findout.

Perhaps that could be clarified by way of correspondence.

I will indeed.

The sheriff——

The sheriffs do a very difficult job. We regard the whole law office, the solicitor, the attachment, the sheriff, the liquidation, as a suite of enforcement powers but we do support our local sheriff.

Given that there were only 689 attachment orders and 23,000 collections by the sheriff, it is apparent that Revenue did not go through the attachment orders in many cases. I respectfully suggest given that Revenue knows the clients and the level of business that if more work was done an attachment order would have been easier to achieve. I encourage more frequent use of attachment orders as opposed to going straight to the sheriff. Given that there has only been 689 attachment orders they have been used in a small minority of cases.

If there was a higher number of attachment orders perhaps the sheriff could operate on a lower number.

Not necessarily, that would not be the main factor. Attachment as a debt collection methodology for Revenue is something we have begun to develop only in recent years. If one looks at the relative figures for 2000 and 2001——

We have three more sections to do.

The details of this may not be reflected in the figures. My understanding is that instead of being selective, the sheriff is given the more difficult cases. For example, if one falls a couple of months behind with VAT payments, the arrival of the sheriff's letter and the visit from the sheriff happen very quickly, almost as a matter of course. I take Mr. Daly's point that the payments often go back directly to the Revenue Commissioners so they may not be accredited to the sheriff for having done the job. That is why the sheriff figures appear to be so large. It is not necessarily true to say there is a selection process where the sheriff gets the more awkward cases. In most instances, the sheriff is the first line of attack. If the payment is not in promptly it is a matter for the sheriff, but the sheriff does not always get the credit for it. Certainly on the ground, that seems to be the mechanism used.

We can clarify those figures. I would be at pains to say it is a suite of collection powers and by and large we try to tailor them to the most appropriate way to collect the debt. Perhaps I can expand on the figures. In 2001, an estimated €100 million was paid directly to the collector general. This was debt that would have been sent to the sheriffs for enforcement and where, as I said, the people went in and paid directly. We should credit that to the sheriffs as well.

Paragraphs 4.8 to 4.10 of the report of the Comptroller and Auditor General read:

4.8 Repayments of Value Added Tax to Registered Traders

Background

Value Added Tax (VAT) is an indirect tax on consumer expenditure that is charged on the value added at each stage of the production and distribution cycle. VAT is chargeable when a taxable person supplies goods or services within the State in the course or furtherance of business. Effectively, VAT is paid by a customer or consumer who is supplied goods or services for his or her own personal use. The supplier (i.e. the registered trader) is charging and collecting the tax on behalf of the Revenue Commissioners. There are currently over 200,000 traders registered for VAT. The gross amount of VAT collected by Revenue in 2001 was €10.5 billion.

A registered trader is charged VAT on the taxable goods and services it purchases for its business. A registered trader in turn charges VAT on the taxable goods and services it supplies. The trader is entitled to deduct the VAT charged on its purchases from the VAT it has charged on its sales. A trader accounts for VAT on his/her sales and purchases by means of a return to Revenue (normally every two months). In most periods, VAT charged on sales exceeds VAT charged on purchases and the trader pays the difference to Revenue. However, in certain circumstances the VAT on purchases exceeds the VAT on sales and the trader is due a repayment of VAT. For instance, if a trader increases stock or purchases an expensive item of equipment, a repayment can arise. Some businesses (e.g. food businesses) are in a permanent repayment situation because their sales are zero-rated but they are being charged VAT on some or all of their purchases. The total amount of VAT repayments by Revenue in 2001 was €2.6 billion. This includes refunds which arise where a trader has overpaid VAT and the overpayment is repaid by Revenue.

VAT is a self assessed tax and registered traders are required to maintain proper records of all transactions that affect their VAT liability. Revenue carry out a programme of audits which includes audits of traders' VAT records to ensure they are complying with the regulations and to check the accuracy of VAT returns including repayable returns. VAT audit activity may be dedicated or form part of general audit programmes. The VAT yield from audits in 2001 was -61.2m mostly arising from underpayment of tax but would also include an element of overstatement of VAT repayment claims.

In recent years, a number of taxpayers have been convicted and received prison sentences for obtaining payment by making false VAT repayment claims. In addition, a former official of the Revenue Commissioners received a prison sentence for conspiring with others to defraud the Revenue Commissioners of €4.8m by means of a false VAT repayment claim.

Revenue's customer service standards aim to repay 85% of VAT claims within 10 working days of receipt of claim with the balance being repaid within a further 20 working days. In 2001, figures supplied by Revenue showed that 80% were refunded within 10 days and 94% within 30 days.

Objectives and Scope of the Audit

The objectives of the audit were

· To ensure that the system for repayment of VAT to registered traders is such that

- Repayments are only made where properly due

- All repayments are supported by valid and complete documentation

- All payments are correctly made and recorded in the accounting records.

· To establish if appropriate arrangements are in place for setting repayment limits and for the approval of repayment claims which exceed those limits.

· To establish whether appropriate management information relating to repayments is produced and utilised to improve the efficiency and effectiveness of the system and its controls.

The examination of the VAT repayments system was based on a review of documentation and discussions with the Accountant General's Office and the Collector General's Office. The role of the Inspectorate in controlling VAT repayments was established by means of discussions with officials in the Office of the Chief Inspector of Taxes. A random sample of VAT repayments and refunds made in 2001 was selected and checked against the prevailing system validation rules. Where repayments in the sample were referred to the relevant tax district for certification the action taken in the district was reviewed.

System for Repaying VAT

In recent years Revenue has been developing its systems so that each taxpayer will be treated as a single customer for all taxes. The Integrated Taxation Processing System (ITP) is a key element in moving to this approach. ITP is a system for issuing and processing returns, payments, repayments and refunds. VAT was incorporated into ITP in April 2000 and since then all claims for VAT repayments to registered traders are processed through the ITP system and payments are made from that system.

VAT returns when received by Revenue are scanned into electronic format and processed through the ITP returns reception sub-system. When the VAT return has been successfully processed through returns reception it will create an amount due from or to the taxpayer for the period depending on whether the return is payable or repayable. Each day the Debits/Credits sub-system of ITP is run which identifies credits on a taxpayer's record and instigates the processing of a repayment. The system checks each repayment claim against a series of validation rules which includes examining the taxpayer's record for outstanding taxes or returns. If a claim fails a validation rule the processing of the claim by the system ceases and a 'work item' is created which is delivered electronically to an official within Revenue to process. If the taxpayer has outstanding taxes the system will, in certain circumstances, automatically set the repayment claim against those taxes. In other situations where there are outstanding taxes, a work item will be created and the repayment claim may be offset against the outstanding taxes manually. The processing of a claim or 'credit' automatically by the computer or manually by an official can result in

· The full amount of the credit being approved and payment being issued.

· The full amount of the credit being disapproved. The credit will remain on the customer's account but there will be a stop to prevent the credit being processed.

· Issuing payment for part of the credit and disapproving the remainder.

· Offsetting the full amount of the credit to outstanding taxes.

· Offsetting some of the credit to outstanding taxes and payment being issued for the balance.

· Offsetting some of the credit and disapproving the balance.

The Debits/Credits system processes the credit through 3 stages

· Credit validation - confirming that the credit is valid and can be released for offsetting and/or repayment.

· Offset validation - offsetting the credit to outstanding taxes.

· Repayment validation - repaying the credit.

Work items created in relation to VAT repayments are referred to the relevant tax district of the taxpayer, to the VAT Repayments Section of the Accountant General's Office or to the Collector General's Office. The official to whom the work item is referred takes the appropriate action depending on the nature of the work item. For instance, if a work item is created because the taxpayer has outstanding returns, the taxpayer will be contacted and asked to submit the returns and informed that payment will be withheld pending their receipt. If a claim fails a particular validation rule and is referred to an official, it is the responsibility of the official to check the claim against the remaining system validation rules, as the claim does not re-enter the sequence of validation checks at the point it failed. The exception to this is a work item referred to the tax district which if approved re-enters the sequence of validation checks at a pre-defined point in the system. An official to whom a work item is assigned can use the comments box provided to indicate what action is being/was taken. The insertion of such comments is optional. When an official has processed a work item to his satisfaction, he electronically approves it and it is then referred to his supervisor for approval. In certain circumstances, for example high value repayment claims, a third level of approval is required.

Audit Findings

Returns

As part of the examination a sample of 40 repayments and refunds made in 2001 were scrutinised. Each payment was checked against the original VAT return submitted by the taxpayer. Prior to scanning, all returns are reviewed to ensure all information appears to be in order. In certain circumstances, for example, if the taxpayer has not signed the return, it will be returned to the taxpayer. Correctly completed returns which have been damaged or which are considered to be uninterpretible to the scanner, are re-written by Revenue staff prior to scanning. In these cases, the original return is attached to the re-written version. This was the case for two of the claims examined both of which were for amounts in excess of €1m.

In seven of the 40 cases examined, the taxpayer submitted a second return for the same period after the original repayment claim had been processed. Each of these cases was a designated 'repayment only case' where the taxpayer is permitted to submit two claims for a period. For other cases, the system will accept a second repayable return for a period and will automatically treat it as supplementary to the first return and not as a replacement for it unless the second return is greater than the original in which case a work item will be created and customer confirmation sought. Revenue should consider introducing procedures to extend the confirmation process to all instances where second repayable returns are received. Alternatively, the VAT return could be revised to allow the taxpayer signify the nature of second returns. The exception to this is where the taxpayer clearly marks the second return as a replacement and this fact is noted by the Collector General's staff prior to scanning. In these cases, the system will substitute the amounts on the second return for those on the first and adjust the taxpayer's record accordingly. This can be problematic if the second return is for less than the first and repayment has already issued based on the first. The taxpayer's record will show an amount due from the taxpayer for the relevant period but this fact is not flagged by the system and recoupment of the amount already repaid will have to await normal collection activities. Revenue has indicated that the possible enhancement of ITP to facilitate the confirmation of all second repayable returns is being examined.

System Validation Checks

The validation checks carried out by the system involve checking against the information held on the taxpayer's record in the Revenue computer. In the main the checks relate to whether the amount of the claim exceeds monetary limits which are set for each taxpayer by the Inspector, whether the taxpayer has outstanding taxes, whether his returns are up to date and whether there are any stops or other markers on the taxpayer's record which would prevent repayment. For the sample of 40 repayments examined, it was found that the system was carrying out all the checks specified.

While staff did receive training and introductory training manuals when VAT was first introduced into ITP, there is a lack of user documentation to inform officials how to deal with work items that arise when claims fail validation checks. However, VAT Repayments Section are currently drafting a user manual. It should be borne in mind that with each new release of ITP, changes can be made to the system that affect the VAT repayments element of the system so that user documentation could become outdated. Nevertheless, it should be possible to maintain an up to date user manual to which staff could refer. Revenue has confirmed that the manual is expected to be completed in October 2002, following which it will be updated regularly.

The VAT Repayments Section is established in such a way that the each taxpayer deals with the same staff member and Revenue's view was that the advantage of this in terms of staff members becoming familiar with the case outweighed the obvious control weakness. Following the audit, Revenue decided to change this policy and will shortly implement the conventional control device of regular rotation of team members. For the sample examined, work items which were referred to officials in the VAT Repayments Section, were dealt with in an appropriate manner. Repayment claims which fail validation rules do not re-enter the sequence of checks at the point which they failed and the onus is on the user to check the claim against the remaining validation rules. Revenue should assess the risk that this gives rise to against the delay in repayment that would occur by subjecting approved work items to re-checking by the system. Revenue has stated that claims which fail validation rules are subject to at least two levels of approval before repayment and that the rechecking of such claims by the system would duplicate work already done and delay the repayment. They are also of the view that even in the unlikely event that the user releases the payment in circumstances where it would fail system checks, the risk is negligible as normal compliance activity would be initiated in due course.

Offsetting Repayments against Outstanding Taxes

The Finance Act, 2000 gave Revenue the power to introduce regulations to allow them to offset repayments due to a taxpayer under one taxhead against outstanding taxes in other taxheads. Previously, this was not possible without the permission of the taxpayer. Revenue introduced these regulations in August 2001. These regulations and the extension of ITP in September 2001 to incorporate almost all taxes with the exception of Relevant Contracts Tax (RCT) and some capital taxes will allow automatic offsetting of repayments due in most instances. The random sample examined showed that repayments and refunds were made in five cases where the taxpayer's record showed amounts outstanding in respect of RCT ranging from €1,825 to €108,146. Two cases were also noted where the taxpayer's record showed Capital Gains Tax outstanding of €14,349 and €21,527. While each of these cases pre-dated the regulations allowing automatic offsetting of repayment against unpaid taxes, they underline the need for the taxpayer's position as a whole to be examined prior to repayment being issued. These cases also highlight the need for Revenue to complete the introduction of all taxes into ITP to allow the system to automatically offset repayments against unpaid taxes. Until this is done, offsetting of unpaid taxes which have not yet been incorporated into ITP can only be done manually and there is the possibility that outstanding non-ITP taxes will be overlooked when repayment is being issued. Repayment claims which do not fail system validation checks and therefore are automatically approved by the system are by definition not brought to the attention of an official. Therefore, it is not possible to carry out manual checks for outstanding non-ITP taxes. It is understood that RCT will be incorporated into ITP before the end of 2002 which will allow greater use to be made of the offset provisions. Capital Gains Tax is now incorporated into ITP and subject to the full offset provisions.

Referrals to Inspectors

The main risk based control of VAT repayments is the system of monetary limits set by the Inspector for each taxpayer. If a repayment claim is received which exceeds these limits the claim requires the approval of the Inspector before payment is made. The main objectives of this system of referral to Inspectors are

· To provide an improved service to compliant taxpayers.

· To assist in the detection of fraudulent or incorrect claims.

In addition, a small proportion of all claims are randomly selected and referred to the relevant Inspector for review prior to repayment.

The setting of these monetary limits is at the discretion of the relevant Inspector. Minimum levels have been prescribed centrally and criteria are set to identify high risk cases in this context, including

· 'Phoenix' type operations or cases with poor Revenue history.

· Traders registering with no fixed place of business.

· Non-resident companies.

· Liquidator/receiver cases.

· Cases where trading records are poorly maintained.

While these criteria are good indicators of potential risks it is essential that cases are constantly reviewed to ensure that all such high-risk cases are promptly identified and the limits adjusted accordingly as necessary.

When a claim is referred to the Inspector it is at the discretion of the Inspector what action should be taken. In essence, the decision has to be taken as to whether the claim can be certified with or without an audit. Instructions to inspectors require that in general all first claims in excess of a specified amount require an audit but that the Inspector has discretion in deciding the degree of checking required. The Inspector may decide that there is sufficient information available to certify the claim otherwise than by way of audit. Factors which the Inspector takes into account in making this judgment include

· 3Whether the case had previously been audited and what is the compliance history.

· What is the nature of the trade and whether it involves highly technical business transactions.

· The degree of understanding by the trader of his/her VAT obligations.

· The agent and his/her level of involvement.

· Are there exempt transactions involved?

· Is there a large volume or value of trade in goods in which the trader does not normally trade or is there a trade in high value goods to registered traders in other EU countries?

Revenue are currently examining the feasibility of establishing VAT Verification Units to deal with the checking of repayment claims. Planning is under way to establish one such unit on a pilot basis. The intention is to review the operation of this unit to establish whether similar units should be set up in all regions.

Of the 40 claims examined during this audit, 17 had been referred to the relevant Inspector for certification because the monetary limits had been exceeded. These claims were certified by the districts on the basis of one or more of the following

· Requesting from the taxpayer supporting invoices, explanation for the claim or other information deemed necessary.

· Knowledge of the taxpayer and their tax history (including knowledge from any on-going audit).

· For single large transactions, checking that the vendor had paid to Revenue the VAT that the taxpayer is seeking repayment of.

· Carry out an audit of the taxpayer.

· Examination of latest accounts.

· Consideration of nature of the business.

In one case where a claim for €1.21m was certified by the Inspector, the comments box on the ITP system in relation to the work item notes that the claim was not checked due to lack of resources. Revenue has since stated that the basis for the decision to approve the claim also included knowledge of the company including a recent audit and the compliance record of the company.

Risk Assessment

In recent years, due to the growth in the economy, the number of claims being referred to Inspectors increased to such a level as to be unmanageable and it became necessary to universally increase the monetary limits for all but the most high-risk traders. In fact, two such general increases have been made in recent years in the level of these monetary limits. These increases were deemed necessary because the level of resources available in tax districts could not deal with the number of claims being referred to them for approval. Nevertheless, the identification of particular risk type cases and the setting of appropriate limits based on the identified risk would be more appropriate than general increases designed principally to reduce the number of claims being referred. The monetary limits are trader specific and therefore are only reviewed if the case is re-examined by the Inspector either because it is the subject of some type of audit or because the Inspector is required to certify a repayment claim. More research and information is required to identify the type of cases that may give rise to the risk of fraudulent or incorrect repayment claims and to apply this knowledge in setting appropriate monetary limits for particular types of cases.

Revenue has recently appointed contractors to develop a computerised risk analysis and risk scoring system to identify customers who are most likely to fail to meet their obligations in respect of the various taxes and duties. The intention is that an automated process of applying rules to available data such as audit results, third party information, accounts, data from tax returns, intelligence information and trade classification (NACE) codes will be implemented to assist Revenue in its audit and investigation activities. This will allow the degree of risk to be determined by the calculation of a risk score. The opportunity should be taken when this system is being introduced to incorporate risk factors in relation to VAT repayments and adjust the system of repaying VAT accordingly. Revenue has stated that the new system will be effective for all taxes and all areas of Revenue business including VAT repayments.

Revenue has estimated that approximately 50% of repayments and refunds are automatically approved by the system for payment. In many cases, a number of years can elapse between audits of traders and therefore it may be some time before the validity of repayments which have not been referred to the Inspector can be checked. Also, when an audit is carried out it does not examine all periods since the previous audit so some repayments may never be examined. In 2001, repayments and refunds of almost €63m were made to traders who registered for VAT in that year. Over €51m of these payments were to companies who according to data on the ITP system have not been audited. Repayments and refunds of over €571m were made in 2001 to traders who have not been audited since before 1990.

Revenue has stated that total reliance cannot be placed on the audit data on ITP as not all districts record audit details on that system. The Active Intervention Management System (AIM) provides a more comprehensive record of audit activity. However, the interface between AIM and ITP systems does not update the record of audit activity. The AIM system is currently under review and the issue of consistent and comprehensive recording of audit activity will be addressed as part of the review. Revenue has also pointed out that some of the companies who received repayments in 2001 with no apparent recent audit were in a group relationship with other companies. While no audit details were recorded in respect of the company that received the repayment, audit activity did in fact take place and was recorded on the computer record of another company within the group. In relation to new registrations, Revenue has stated that they are aware of the risks involved and have procedures in place to minimise that risk. These procedures include checking the PPS numbers of the principals involved in the business, examining that compliance record of associated businesses and in certain cases conducting pre-registration visits.

In relation to VAT risk analysis, Revenue stated that

· The increase in the referral limits followed a review in 2000 by a working group in the Chief Inspector's Office. It was a necessary stop gap measure to alleviate an unacceptably high level of non-productive referrals and to provide a breathing space in which to develop a real risk based selection model.

· Various options for the development of a more scientific method of case selection had been examined in recent years including

- Flexible risk ratings based on industry sector, case size, compliance record, etc.

- Differentiating between 'payment' and 'repayment' traders so that different targeting rules could be applied to each category

- The use of audit dates so that traders who had been satisfactorily audited could be given higher referral limits.

· In the absence of an overriding risk analysis system, the Chief Inspector's Office has centrally identified and issued to selected tax districts lists of VAT cases which posed a possible revenue risk. These were based on factors such as repayment only cases, cases in an overall repayment position and cases that had received two repayments in excess of the monetary limits in a calendar year.

Revenue has also stated that risk, in the context of Revenue operations, focuses on the potential loss of revenue arising from non-compliance with obligations under legislation with the purpose of identifying those who pose the greatest risk of tax default. In this context risk is not confined to repayments only.

Management Information

Possible risk based checking of claims is of course limited by the amount of information that is available. As the VAT return shows the amount of VAT on sales and purchases for the period but not the amount of sales and purchases themselves, it would not be possible to review ratios. Registered traders are required to submit each year a return of trading details showing the total of their sales and purchases for the year broken down by VAT rate. These returns are required to be submitted to the Collector General's Office but due to a lack of resources the information on returns submitted is not being captured on the taxpayer's record. This limits the ability to check the credibility of repayment claims against known trading levels. It is understood that proposals to resolve the matter are currently under consideration.

Each taxpayer registered with Revenue is given a trade classification code known as a NACE code. Proper classification of traders by type of business would help in assessing the validity of repayment claims. However, the current system of NACE codes is apparently not suitable for this due to a lack of definition in the code. A project is under way to review and revise the NACE codes.

Other than information in relation to the number and value of repayments made and performance in relation to customer service standards, very little management information is routinely produced in relation to VAT repayments. The production and review of relevant management information would provide a valuable tool in monitoring and controlling VAT repayments. Examples of the type of management information that could be obtained include

· Analysis of claims disapproved or partially approved by Inspectors. This could provide indicators of possible fraudulent claims and inform checks on future claims. Further analysis could be conducted by trade class to identify types of traders whose claims are adjusted most often.

· The random selection of claims for review by the relevant Inspector is capable of providing a valuable indicator as to whether the system is operating satisfactorily. However, unless the results of the Inspectors' reviews of these claims are collated and analysed, an assessment of the system based on this test can not be made.

· Details of claims from recently registered traders.

· The yield in terms of payment or offsetting of outstanding taxes from checking claims should be recorded.

Revenue has stated that the review of the NACE codes and the new computerised risk analysis system will allow greater analysis of available information for management information purposes.

Conclusions

The administration of VAT differs significantly from that of other tax heads by virtue of the fact that almost one quarter of the gross amount collected is subsequently repaid. This is innate to the nature of the tax but it requires Revenue to continually reassess its strategies for striking the correct balance between legitimate demands for improved customer service in making VAT repayments with the equally important objectives of only making correct repayments and minimisation of the likelihood of fraud.

The consolidation of VAT, including repayments, with the integrated taxation processing system in April 2000 provided the opportunity for significant benefits both in the areas of customer service and internal control. However, the relative newness of these arrangements requires ongoing strategic and operational review to ensure that the full possibilities of the system are achieved and that where necessary it is modified to address any weaknesses detected.

The system now in operation puts a premium on the implementation and regular review of a comprehensive risk control strategy, together with full operation of the procedures designed to achieve the desired level of control. The findings of my examination indicate that there is scope for improving both the strategy and the application of controls. Revenue is now at an advanced stage in obtaining a comprehensive computerised risk management system for all taxes and all areas of Revenue business including VAT repayments.

There is an increased level of risk associated with the issue of substantial repayments to newly formed companies and to older companies which have not been audited for some time. Revenue has confirmed that with 25,000 new VAT registrations annually, risk analysis and risk management must be central to the control process. The tax history of the individuals behind each new business are checked in detail and further action such as new business visits and later audits are based on the on going assessment of risk. However, my examination noted low utilisation of management information for control purposes in an area where the volume of transactions would suggest that it could be a valuable control tool.

Potential weaknesses at the point where manual intervention necessarily allows for system checks to be overridden should be carefully managed, for example where there is a break in the sequence of validation tests. Clarity in the handling of second returns for a period, the better use of comments fields, improved user documentation, a more refined use of business codes together with the implementation of the staff rotation policy should help to further improve the control environment.

4.2 Dividend Withholding Tax

Background

The scheme of Dividend Withholding Tax (DWT) was introduced in the Finance Act, 1999, and requires companies resident in the State to make a deduction at the standard rate of income tax from dividends paid or other profit distributions made after 6 April 1999, subject to certain exemptions. Recipients such as companies resident in Ireland, investment funds, charities and pension funds are specifically excluded. Others, including most non-resident individuals and companies, may qualify for an exemption by making the appropriate declaration and supplying adequate supporting documentation either directly to Revenue or to approved dividend paying companies or agents acting for the company (Authorised Withholding Agents). A financial institution or stockbroker which has registered with Revenue as a Qualifying Intermediary may receive distributions without deduction of DWT in respect of clients who have provided the intermediary with appropriate declarations of exemption. The Qualifying Intermediary takes responsibility for administering the DWT, must retain declarations and other relevant documentation for inspection for six years, and provide an auditors report of compliance with the scheme in its first year of operation and subsequently on request.

A company or Authorised Withholding Agent which makes a distribution is required to submit a return containing details of distributions made and DWT deducted to the DWT Section of the Customs and Residence Division of Revenue by the 14th day of the following month, together with payment of the full DWT liability. A return is required even where no DWT deduction is made from the distribution. An Inspector of Taxes may make an assessment for any unremitted tax or in cases where s/he considers that the return understates the full amount due. While the recipient of the distribution is required to declare the gross amount received for income tax purposes, credit may be claimed in respect of the DWT deducted at the standard rate. Where the DWT amount exceeds final tax liability, or where a valid exemption entitlement had not been sought, a refund of DWT may be claimed.

DWT receipts for 1999-2001 are shown in Table 1.

Table 1 - DWT Receipts and Returns 1999-2001

Gross Receipts

Refunds*

Net Receipts

Total No. of Returns

No. of ’Nil’ Returns

1999

€46m

€8m

€38m

1,003

378

2000

€203m

€18m

€185m

3,301

1,429

2001

€189m

€46m

€143m

4,010

1,628

[* The table shows the overall DWT position, taking account of direct refunds by DWT Section and tax districts, and subsequent refunds on the basis of approved claims from non-residents etc.]

The low figures for 1999 reflect the nine month period, and special transitional arrangements for some non-resident persons. It was also likely that some distribution dates may have been brought forward to avoid the necessity for a DWT deduction. Revenue expectation was that DWT would yield in the region of -100m in a full year.

Objectives and Scope of the Audit

The objective of the audit was to examine the extent to which procedures currently operated by Revenue provide adequate assurance as to the assessment and collection of revenues due in respect of DWT. Audit work mainly focused on the operation of the DWT Section. Information and papers were also received from the Office of the Chief Inspector of Taxes. From an examination of records, a sample of recipients of distributions were selected and cross-checked to their IT returns. A sample of companies was selected from those which had previously made distributions for Advanced Corporation Tax purposes and failed to submit a DWT return. The sampled cases were cross-checked to Corporation Tax returns and company accounts for evidence of distributions. Procedures for granting exemption status from the payment of DWT, and for the issue of refund payments, were also examined.

This report also includes reference to a number of findings from a 2001 report by Revenue's Internal Audit Division on the procedures governing the administration of DWT.

Audit Findings

Declaration of Distributions by Companies

DWT operates on the basis of self-declaration by companies. Companies are not required to register for DWT, and the Revenue returns compliance programme is not applied to company distributions. There were no comprehensive procedures operated to ensure that all liable companies submitted returns for DWT. Corporation Tax returns and supporting annual financial statements were not systematically monitored for distributions which would indicate a liability for DWT returns and possibly deductions. Revenue has however stated that an enhancement is in the process of being developed whereby the response to the panel on the Corporation Tax return indicating whether a distribution had been paid will be captured by the Tax Districts and made available electronically to the DWT Section.

The level of returns received in 2000 and 2001 of 3,301 and 4,010 respectively (relating to some 2,300 companies) represents under 4% of companies classified as live for Corporation Tax purposes. For the same periods, DWT returns were submitted by only 50% of those companies which had returned Advanced Corporation Tax in 1997/98. Revenue have pointed out that the corresponding figure for 1998/99 rose to 57%. While it considers that these statistics in themselves do not indicate a DWT compliance problem, Revenue accepts the need to crosscheck former Advanced Corporation Tax filers against DWT filers and the DWT compliance programme now provides for this.

A review by my staff of 21 companies which had filed returns in respect of Advanced Corporation Tax but which appeared not to have submitted DWT returns revealed that

· 13 companies had not made distributions; however in five cases, distributions may have been made by other companies in the group

· 5 of the companies had, in fact, made returns in respect of the distributions identified. One of the companies filed a return in 2000 but had omitted to file a return in 1999. The liability was €7,807 and the return has since been received

· 3 companies had made distributions totalling €1.4m without completing a DWT return. Two of the companies had no liability and are now in the process of preparing 'nil' returns. In the other case, confirmation is awaited as to whether a proposed distribution was made.

Internal Audit noted that 439 or 25% of companies which had made a DWT return in 1999/2000 did not submit a return in 2000/01. Revenue stated that there may not be a compliance problem as a company which makes a distribution in one year may not do so in the following year. However, in June 2002, 79 companies which had previously filed a return in 1999 and 2000 showing a DWT liability in excess of €1,000 but did not file a return in 2001 or early 2002 were contacted. To date, 38 companies had responded and only 2 had a DWT liability which amounted to €11,000 in total.

Internal Audit found that 25 of a sample of 79 companies which had made Corporation Tax payments greater than €125,000 had failed to submit DWT returns. The formal response to Internal Audit stated that, of the 25 cases,

- 16 companies should have submitted 'Nil' returns

- 1 company had a DWT liability and subsequently filed a return

- 1 company had made the distribution prior to the introduction of DWT

- 2 companies which had provided for distribution in their accounts did not make the distributions

- Investigations were under way into othe remaining 5 companies.

Revenue has pointed out that 92% of DWT yield comes from stock exchange quoted companies plus a handful of others (111 companies in all) - and these are very tightly controlled. In the case of companies quoted on the Dublin Stock Exchange, a weekly list of companies, who are due to make distribution is provided by the Dublin Stock Exchange to DWT Section. These companies are monitored to ensure that returns and payments are received on time. The compliance level has been excellent and ensures that the bulk of DWT is collected in full.

Revenue stated that the introduction of a comprehensive compliance programme for smaller companies was postponed until 2002 as it was considered that any loss of revenue would be very small, and that the initial focus had to be on customer service and on establishing a system to administer the tax. Revenue also stated that DWT Section has commenced a compliance programme to gauge the level of compliance of the smaller private companies, and that the only evidence of a compliance problem to date related to companies which did not have a DWT liability but which failed to file the required 'nil' return.

Claims for Exemption Status

A sample of exemption cases was examined by my staff and all appeared to be in order and were supported by appropriate declarations from the beneficiary. Variations were noted in the level of assurance provided in the audit certificates supplied by Qualifying Intermediaries. In this regard, it is noted that, following a consultation process with the accountancy profession, DWT Section had agreed a standard certificate with the objective of achieving a consistency in the audit work to be performed and the extent of the assurance to be provided. In addition, the Section has drawn up a set of audit checks for the examination of Qualifying Intermediary records, and a programme of Revenue visits to these agents commenced recently.

Regular Distribution Patterns

There are strong indications that salary payments and other remuneration are being made by companies in the form of distributions which initially only attract deductions at the standard rate of tax, and which also avoid the requirement for payment of PRSI contributions. The Internal Audit report noted 32 companies which had made monthly distributions of similar amounts to the same beneficiaries. The DWT Section has since identified 139 companies making regular distributions. The matter was initially referred to Special Enquiry Branch for investigation but Revenue has since referred the cases in question to its Anti-Avoidance Unit which will handle such cases in future.

Declaration of Dividend Income on Income Tax Returns

While DWT is deducted at the standard rate from the proceeds of distributions received, a taxpayer has an Income Tax liability at the marginal rate. However, while the name and address of recipients must be shown, there is no requirement to record the Personal Public Service (PPS) No. of dividend recipients on DWT returns. Even in the absence of this common reference, it is still possible to perform a manual matching between DWT and Income Tax returns for individual cases but an overall systematic check on the return of dividends for Income Tax, possibly through the selection and sampling of high value cases, is not feasible.

Revenue stated that, at the time the DWT legislation was enacted, it was considered that imposing a requirement on registrars of quoted companies and qualifying intermediaries to obtain the PPS numbers was a disproportionate compliance burden which could interfere with the efficient workings of the stock market. Revenue said that a recent report recommended undertaking a study to establish the benefits accruing and costs associated with the quoting of third party reference numbers on all forms, including the DWT form. The results of this study will inform the debate prior to requesting any legislative changes to implement the quoting of such PPS Nos. Revenue also stated that a prototype system for better matching and risk profiling of dividend recipients has been developed and will be made available shortly to Tax Districts.

A sample of 50 recipients of dividend income to the value of -1.6m was extracted by my staff from DWT returns and matched with individual Income Tax returns with the following results:

· 33 returned the dividend

· 5 returned an amount less than the dividend paid on DWT return

· a PPS No. could not be traced in 4 cases

· 2 did not submit an Income Tax return

· in 6 cases the submitted Income Tax returns were not traced by Revenue; a computer check has established that the relevant Income Tax returns have been processed in 5 cases, assessments have issued and that each has returned dividend income. The remaining case is under enquire by the tax district.

Internal Audit found that, of 17 Income Tax returns examined, 5 individuals had not returned their dividend. Revised returns and tax payments were subsequently received from 3 individuals. A further case, while not returned by a minors' trustee, was returned on the minors' returns. The final case was resolved by a repayment of the DWT.

Conclusions

One of the lessons from the DIRT investigations was that, even in a self-assessment system, it would be inadvisable to rely totally on self-declarations by companies, or any other clients of the tax system. There are clear indications both from the findings of this audit and from the Internal Audit report that compliance by companies with the DWT scheme is not being monitored by Revenue in a systematic way.

The DWT Section is now becoming more proactive, for example by launching a programme of checking of recipients' agents. It is likely that this process will be accelerated by the Internal Audit report. Nevertheless, up to recently, DWT was less than fully established in the tax system, with the level of compliance unclear. It is considered that further improvement in this situation would be achieved with actions on two fronts:

· setting DWT compliance targets and implementing a clear strategy to achieve them, and

· putting structures in place to facilitate full exchange of data with other areas of Revenue.

Part of the administration of DWT is delegated to companies, to Authorised Withholding Agents and to Qualifying Intermediaries. While this makes a substantial contribution to the efficient operation of the scheme, Revenue should formally consider the risks inherent in a delegated approach. An appropriate programme of monitoring of this aspect of the scheme should then be implemented which is commensurate with the level of risk and the efficient utilisation of resources.

In response to the report, Revenue has stated that since the introduction of the scheme, much of the emphasis was on legislative, customer and information technology development issues. As priority was now switching away from those issues, greater emphasis and resources were being assigned to scheme compliance issues. Revenue has indicated that actions now under way or proposed include:

· 14 reviews are currently in progress under a programme for reviewing and checking the activities of paying companies and intermediaries. It is envisaged that there will be a significant expansion in this aspect of work in the future

· since September 2001, DWT is in the Integrated Taxation Processing system and, since May 2002, returns and payments can be filed through the Revenue On-line System. A comprehensive DWT database containing details of DWT payments and related shareholder information has been developed and is available to DWT section and is being rolled out to tax districts

· DWT Section intend to compile, on an ongoing basis, a register of companies with a potential DWT liability. Previous filers will be recorded and any who have not made a return in the previous 18 months will be contacted

· as part of the compilation of a DWT register, former Advanced Corporation Tax filers will be cross checked against DWT filers, and the position of all companies who have not made DWT returns will be checked

· electronic transfer of data from the Corporation Tax returns to DWT Section

· Corporation Tax filers who had a tax liability greater than €125,000 and who did not make a DWT return will be identified, and the accounts submitted to tax districts in these cases will be inspected.

4.2 Forecasting of Tax Receipts

Under Article 28 of Bunreacht na hÉireann, a White Paper setting out Estimates of Receipts and Expenditure for the coming year is presented to Dáil Éireann prior to the annual Budget statement of the Minister for Finance. The estimated tax receipts in this statement are subsequently adjusted to take account of Budget changes, and the resulting 'Post-Budget Estimate' becomes the tax collection target for the coming year, and the main funding basis for the expenditure side of the Budget statement. While the final estimates of tax receipts are decided and formally presented by the Department of Finance, and while the estimates at all stages are prepared on the basis of economic forecasts from the Department, Revenue plays a major advisory role through the preparation of estimates of tax receipts and costing of options and proposals. Tax receipts are reported through the monthly Exchequer Statement, the annual Finance Accounts and my Annual Report (which differs from the other sources in showing the actual net receipts as opposed to the amounts paid into the Exchequer).

Table 2 - Variation between Forecast Tax Receipts and Exchequer Outturn 2001

Tax Head

2001 Post-Budget Estimate €m

2001 Outturn €m

Actual Shortfall €m

Shortfall As Percentage of Estimate %

Customs

230

165

-65

-28

Excise

4,774

4,050

-724

-15

Capital Taxes

1,196

1,051

-145

-12

Stamp Duty

1,276

1,227

-49

-4

Income Tax

9,879

9,347

-532*

-5

Corporation Tax

4,302

4,156

-146

-3

VAT

8,792

7,920

-872

-10

Other

12

9

-3

Total

30,461

27,925

-2,536

-8

*includes PAYE shortfall of €762m.

The considerable variances which occurred in 2001 between the final forecast of tax receipts for that year and the actual amounts collected for each tax head and paid into the Exchequer are detailed in Table 2. The overall outturn for 2001 was 8% less than predicted, and there were notable shortfalls in VAT, Income Tax and Excise. The outturn for Excise for 2001 was €213m lower than the 2000 total in a year when the preliminary estimate for GDP economic growth was just under 6%.

In the current year, the end-March 2002 Exchequer returns noted that year on year tax receipts were 2.8% lower in the first quarter of 2002 than for the equivalent period in 2001, as compared with a predicted increase of 8.6% for all of 2002 over 2001. Internal Revenue statistics for 2002 Income Tax for the period from 1 January to 31 May 2002, which are analysed in Table 3, indicate a shortfall of €295m (7.7%) over expected receipts in the period. The cumulative shortfall in PAYE, which is the major component of the Income Tax total, was €182m (5.8%).

Table 3 - Income Tax: Projected and Actual Receipts to end May 2002

Projected*to end May 2002 €m

Actual Receipts**To end May 2002 €m

Shortfall €m

Income Tax (PAYE)

3,126

2,944

182

Income Tax (Non-PAYE)

695

582

113

Income Tax (Total)

3,821

3,526

295

*For management information purposes, the Post-Budget Estimate for each tax head for the year is apportioned by Revenue over each of the months on the basis of the established payment patterns for that tax head.

**Internal Revenue collection figures are not directly comparable with Exchequer lodgments.

Having regard to the serious implications for the economy of inaccurate revenue forecasting, I asked the Accounting Officer

· whether specific analysis has been undertaken with a view to establishing the factors which have given rise to the 2001 shortfalls in VAT (€872m), Capital Taxes (€145m) and Excise (€724m) and, if so, the results of such analysis

· the reasons for the continuing trend of shortfalls in the PAYE take, and if these shortfalls are mirrored in the PRSI take for 2001 and 2002 to date.

In addition, I sought the observations of the Accounting Officer as to the extent to which the recent variances in revenue collected as against that forecast, and in particular those outlined above, are indicative of

· a weakness in Revenue forecast data, calculations or methodology

· a weakness in Department of Finance forecast data, calculations or methodology, or

· a reduction in the efficiency and effectiveness of revenue collection.

I also invited the observations of the Accounting Officer at the Department of Finance.

Since I received the observations of the Accounting Officers the shortfall between projected and actual income tax receipts has grown further as shown in Table 4.

Table 4 - Income Tax: Projected and Actual Receipt to end August 2002

Projected to end August 2002 €m

Actual Receipts To end August €m

Shortfall €m

Income Tax (PAYE)

4,929

4,566

363

Income Tax (Non-PAYE)

863

729

134

Income Tax (Total)

5,792

5,295

497

Revenue Response

In response the Accounting Officer of Revenue pointed out that the role of Revenue in forecasting tax revenue receipts was a supporting one to the work of the Department of Finance. He stated that each year Revenue was formally asked by the Department of Finance for preliminary forecasts of tax revenue. The forecasts for some taxes such as PAYE, VAT, Capital Gains Tax, Customs, and the VRT element of Excise Duties were normally derived by Revenue from applying the percentage growth rate projected by the Department for certain macro-economic indicators to the expected tax outturn of the preceding year. Appropriate specific macro-economic indicators were not available for other taxes such as Corporation Tax, Income Tax (non-PAYE) and Excise (other than VRT) and forecasts for these were based on alternative growth indicators derived from local or other sources. While the specific macro-economic indicator used for VAT, namely Personal Consumer expenditure, could also be used as a basis for forecasting Excise, other than VRT, it was considered much less appropriate because the Excise base was significantly narrower than VAT. Accordingly the prevailing method for forecasting Excise was by way of a trend-based analysis of the yields from the various dutiable commodities over a number of years. The forecast figures thus provided by Revenue were taken into account by the Department of Finance in their deliberative process leading up to publication of the Pre-Budget forecasts (the "White Paper" Estimates) and the Post-Budget forecasts. The latter reflected the projected impact of Budget changes on the economy as well as on tax revenues.

Specific Analysis

As regards specific analysis, the Accounting Officer informed me that in 2001 the collection performances of all taxes were monitored as usual on a monthly basis and that the forecasts were continually revised where appropriate until year end. A number of special enquiries were also undertaken during the year into the collection performances of VAT, Excise and PAYE which were significantly under-performing against Budget forecast.

With regard to the factors giving rise to the shortfalls in VAT, Capital Taxes and Excise the Accounting Officer stated that, of the total shortfall of €872m for VAT against Budget target, VAT on Imports accounted for €233m with the main causes of the undershoot being a significant drop-off in imports of motor vehicles, fuels and computer parts from outside the EU. While falling significantly short of Budget forecast, the yield in 2001 from VAT Internal grew by over 8% compared with the outturn for 2000. That was a marked fall in the growth rate of 18% achieved in the year 2000 over 1999. Revenue was satisfied that the reasons for the fall in the rate of growth in 2001 was directly related to the level of activity in the economy reflecting a general slowdown in consumption, exacerbated by the Foot and Mouth crisis and influenced in the last quarter by the attacks in the USA on 11 September. Because of the general fall in growth rate throughout 2001, it was not possible to place monetary values on each of these effects.

The Accounting Officer said that Capital Taxes were more difficult to forecast than other taxes generally because of their nature. Capital Gains Tax and Capital Acquisitions Tax were driven by events (mainly disposals of assets), and were not as closely related to general economic activity as VAT or Excise. Of the total shortfall of €145m for all Capital Taxes against Budget target, Capital Gains Tax accounted for €113m. He said that firm reasons for that shortfall were not available but that, while an element of it could have been attributed to some slowdown in the economy in early 2001, it may also have been an indication that the release of pent up investment funds which generated the acquisition of further capital assets in the years following the reduction of the Capital Gains Tax rate to 20% in the 1998 Budget had slowed down. The shortfall of €32m in yield from Capital Acquisitions Tax was accounted for principally by a reduction in the yield from Inheritance Tax.

The Accounting Officer also stated that receipts from Excise were down on target for all excisible commodities. An examination of the net receipt in 2001 indicated the following breakdown of the €724m shortfall as identified at commodity level in Table 5:

Table 5 - Excise: Analysis of 2001 Shortfall by Commodity

Commodity

Shortfall (€m)

Commodity

Shortfall (€m)

Beer

63

Tobacco

188

Spirits

57

Light Oils

66

Wine

14

Other Oils

56

Cider & Perry

3

VRT

276

The total of these figures emerges at €1m less than the total shown in Table 3 because of rounding in the constituent items.

While collection performance in general reflected a fall in consumer spending on excisable items, he added that some specific factors were also identifiable such as:

· the provision in the 2001 Finance Act to abolish the end-year payment catch-up for alcohols was not included in the Budget arithmetic - the cash flow cost of this factor at end 2001 was estimated at €39m for Beer, €33m for Spirits and €16m for Wine

· the negative effects on yields arising from the Foot and Mouth crisis, particularly in the tobacco and oils sectors (these were down on target by €188 million and €122 million respectively)

· the yield from VRT was affected by a drop of more than 66,000 new car registrations in 2001 compared with 2000 and with a 2001 Budget forecast which was based on an assumed "no change" year-on-year in the volume of car registrations (this was down on target by €276m).

The Accounting Officer informed me that the shortfalls in PAYE in 2001 and in 2002 to date were the subject of ongoing research. Various aspects of Income Tax forecasting and collection, particularly in regard to PAYE, were analysed on an ongoing basis by a technical group chaired by the Department of Finance and involving personnel from Finance, Revenue, Social and Family Affairs and the Central Bank. He indicated that some specific contributory factors towards the €762m shortfall in 2001 PAYE collection as against Budget forecast had been identified, such as an over-estimate of €67m in the 2000 outturn base when compiling the 2001 forecast, an underestimate by €203m in the cost of the 2001 Budget package in 2001 and the cost of the Special Savings Incentive Scheme, which was introduced in the 2001 Finance Act and not factored into the Budget arithmetic. The cost of the Savings Scheme in terms of PAYE tax forgone in 2001 was €55m but when that was offset by a reduction in the expected cost of Medical Insurance relief under the new Tax Relief at Source system the net cost was reduced to €27m. In the wider economy, the lower than expected growth in GDP in 2001 allied to a reduced growth in employment numbers (lower at 49,000 than the 60,000 estimated at Budget time) would also have affected income tax.

The Accounting Officer considered that most taxes were performing reasonably well in 2002 apart from Income Tax. As regards the negative collection performance by PAYE in 2002 to date, Revenue was confident that part of the variation was explained by issues of timing which could not be accurately predicted because of changing the tax year to the calendar year, and the difficulty of predicting exactly when budget changes would take effect. Also, the monthly targets for PAYE collection in 2002 were broadly based on the monthly collection pattern over the 12 months of 2001. PAYE collection in 2001 was at its strongest in the first half of the year and using this as a basis for the corresponding target in 2002 may have had the effect of front-loading the target for the first half of 2002 to an unrealistic level. The 2002 forecast for PAYE was based on a projected growth over 2001 allied to an expected recovery in the economy. If this recovery takes place, PAYE receipts should recover over the remainder of the year in line with the general pattern in the economy.

While Revenue did not monitor PRSI collection against monthly targets, the Accounting Officer understood that PRSI outturn had come in ahead of targets set by the Department of Social and Family Affairs in 2001 and to date in 2002. As regards the comparison between PRSI and PAYE collection performance there were certain systemic differences in the way PAYE and PRSI operate which can account for different effects in the respective outturns. For example, if earnings were reduced (short-time working, less overtime or job loss) there was no refund of PRSI contributions already paid but there could be tax repayments due. Furthermore, any reduction in earned income for individuals who had already exceeded the PRSI ceiling (€35,870 in 2001 and €38,740 in 2002) had no effect on the individual's contribution to PRSI but it would reduce the tax yield.

Quality of Revenue Forecast Data, Calculations and Methodology

The Accounting Officer informed me that the forecasting of tax revenue could not be an exact science as it was necessarily based on a number of key factors which were themselves forecasts. Tax revenue forecasts were usually based on a combination of forecast yield figures for the outgoing year (or for a future year, where forecasts for more than one year were being prepared) and on projected annual trends in economic growth. Frequently it was also necessary to factor in an estimated carryover cost or yield impact of significant budgetary changes, which impact had itself been compiled on the basis of historical data projected forward using extensions of the same projected trends in economic growth. In summary, forecasts of tax revenue would be as sound or as weak as the variables on which the process depended and that while every effort was made to have reliable basic data there would also be scope for some margin of error. Tax receipts performance in 2002 was being kept under review by the technical group.

Possibility of Reduction in Efficiency and Effectiveness of Revenue Collection

The Accounting Officer assured me that Revenue had examined in considerable detail the question of whether the extensive shortfalls were indicative of a reduction in the efficiency and effectiveness of collection, and was fully satisfied that there was no such reduction. The pattern of return and payment compliance for PAYE is shown in Table 6. Similar patterns pertained to VAT and other taxes.

Table 6 - PAYE Compliance: Number of Employers paying PAYE on time

Year

Number of monthly returns issued to employers

Number of monthly returns/payments on time

Percentage compliance*

1999

1,373,321

702,069

51.12%

2000

1,534,609

805,224

52.47%

2001

1,676,013

896,963

53.52%

2002 (to end-May)

718,785

404,992

56.34%**

*The figures shown related to returns received within the due month. For example in 2002 by the end of the following month, 74% of all cases which had a known liability had made returns.

**Although percentage compliance appeared to be low (56% in 2002 to end May) the proportion of monthly liability represented was much higher as it included the big cases and most medium sized payers.

Another measure of compliance was the level of arrears outstanding. He stated that there had been no exceptional rise in the level of arrears at December 2001. Arrears of €351m at that date represented a rise of 1.4% over December 2000, while collection rose by 2.7%. The final possible area in which there might have been a reduction in compliance would be the area of underdeclaration and, in particular, whether the extent of undeclared employment increased in 2001. Any significant rise would be visible as a divergence in trend between the number of employees and employments reported to Revenue, and the general employment statistics. The Quarterly Labour Force Estimates for 2001 showed a year-on-year rise in the number of persons in full-time employment of 3.6% at the start of 2001, falling to 2.5% in the final Quarter. Returns received by Revenue in 2002 (in respect of 2001) showed an increase of 7% in the number of employments. While the significance of the larger increase in the number of employments was under investigation, Revenue was satisfied that there had been no widespread failure to declare employment. In conclusion, the Accounting Officer indicated that Revenue was fully satisfied that taxpayer compliance and Revenue efficiency had been maintained or enhanced since the beginning of 2001, and had made no contribution to the relatively poor performance of that year.

Response of Department of Finance

Economic Growth Lower than Anticipated

In response to my request for observations, the Accounting Officer at the Department of Finance informed me that the shortfall in tax receipts was mainly due to overall economic growth being lower than expected on Budget day. Budget 2001 was predicated on a forecast of 8.8% for real GDP growth for the year 2001. That economic forecast was based on the latest national and international data available to the Department at that time and was broadly in line with what was widely believed as the likely growth scenario for the economy. The EU Commission Autumn 2000 forecasts expected real GDP growth of the order of 8.25%, the OECD and the Central Bank about 8% and market commentators' forecasts did not differ substantially from that level.

However, the outturn for economic growth was much lower than anticipated, with the preliminary estimate for GDP growth for 2001 at 5.9%. That lower growth was due to the outbreak of Foot and Mouth disease and related restrictions which also affected the pattern of growth (depressing cross-border and domestic spending on petrol, for example), as well as to the effect of the slowdown in the United States economy, exacerbated by the events of 11 September. That lower economic growth and changed patterns of personal consumption generated less tax receipts than expected, especially in the areas of Income Tax, Value Added Tax and Excise. In the case of excise receipts €88 million of the shortfall was due to an oversight, as the cost of deferment of the payment of excise duty on alcohol was not included in the Budget arithmetic.

Tax Forecasting Methodology

The Accounting Officer stated that the tax forecasting methodology used by the Department was set out in a report published by the Department in 1999. Basically, the process of forecasting involves collaboration between the Revenue Commissioners and the Department. In the run up to the Budget, the Revenue Commissioners provide estimates of likely tax revenue for the Budget year in question and the subsequent years. In doing so, the Revenue Commissioners rely in certain cases (e.g. income tax) on macroeconomic forecasts supplied by the Department of Finance on earnings, employment and consumption growth. In assessing the overall tax revenue forecast the methodology involves relating that forecast to expected economic growth in nominal terms. The relationship is of the order of 1:1 with variation to be expected, depending on whether growth is mainly domestic or externally driven, and on whether budget tax measures are less or more generous. If growth is mainly externally driven or tax concessions are more substantial then when nominal GDP increased by 1%, taxes would be expected to increase by less than that.

In the period 1996 to 2000 the ratio (or elasticity) of the increase in taxes compared to nominal GDP averaged close to 1:1 i.e. every 1% increase in actual GDP resulted in an increase in taxes of 1%. The 2001 Budget elasticity based on the 2000 forecast out-turn for taxes was 0.84 and 0.91 based on the actual 2000 outturn. The forecast for the increase in tax revenue in the Budget for 2001 was 11.6%, which became 12.5% as the tax out-turn for 2000 came in €216m below the Budget forecast.

Factors which Impacted on Income Tax Yield in 2001

The Accounting Officer considered that at the time of the Budget for 2001 the forecast increase in tax revenue appeared reasonable and consistent with previous experience. He said, however, that the experience in 2001 had turned out much differently. One relevant factor had been the decrease in average hours worked in the economy in 2001, which in the industrial and construction sector amounted to one hour or more. That in itself could have been expected to have a significant effect on the rate of growth in income tax receipts.

In the case of income tax, the forecast tax yield was further affected by an upward revision in the cost of the Income Tax reductions in the Budget for 2001 on the basis of subsequently available data. The annual update of the tax base completed in summer 2001 led to an upward revision in the cost of the tax package which had been announced in the Budget. The revised tax base reflected the fact that economic growth in 2000 was higher than expected in late 1999, leading to higher employment and incomes in 2000. For example, the Department of Finance forecast for GDP and GNP in the 2000 Economic Review and Outlook was 10.3% and 8.3% respectively. The eventual figures published by the CSO for the year 2000 were 11.5% and 10.4%. The revised tax base reflected the fact that the estimated number of PAYE income earners on Revenue's books in 2001 increased by 76,000 compared with the corresponding figure used in the Budget for 2001. The tax base revised on the basis of these figures caused the cost of the 2001 tax package to be revised upwards by €300 million in a full year (€200 million in 2001, i.e. April to December tax year).

A further consideration was that tax system changes (e.g. the widening of the spouse tax band, broadening of tax relief on pension contributions) could affect behavioural patterns, with the potential to affect tax yield in ways in which there was no "history" on which to ground estimates of impact. He said that substantial changes in the tax regime of recent years may have generated effects of that kind, the impact of which would only be determinable as later years' tax data became available for analysis.

Factors Impacting on Tax Yield in 2002

In relation to 2002, the Accounting Officer reported that tax receipts to end-June were down 7.1% compared with receipts during the same period last year. However, he pointed out that the due date for June Corporation Tax had fallen on Friday 28 June, and that payments of the order of €1 billion had been received on 1 July in relation to the June period. An adjustment for that timing factor left tax revenue at roughly the same level as 2001.

He indicated that tax receipts for the first half of 2002 had been affected by a number of factors:

· The tax reductions in the budget for 2001 came into effect on 6 April 2001 and there was a carry over cost of the full year effect of those reductions in the first four months of 2002. That cost was estimated at €540m in the first four months of 2001.

· The start of the income tax year had been brought back from 6 April to 1 January. As a result, the tax reductions announced in the Budget for 2002 were payable with effect from 1 January 2002, and not from 6 April as was the case in 2001. The additional cost of that was approximately €120m.

· Refunds of income tax of €170m were made in respect of Special Savings Investment Accounts (SSIAs) in the first six months of 2002 compared with €2m in the same period in 2001. SSIAs were introduced with effect from 1 May 2001.

Income tax receipts at the end of June 2002 were €630m below receipts at the end of June 2001, whereas the total of the foregoing measures was €830 million. He pointed out that this year's yield was also affected by increases in income over the levels obtaining in 2001 ^ the size of which was conjectural at this point.

He also noted that the economy was growing at a much faster rate in the first half of 2001 compared to what was expected for the first half of 2002. That was a result of the sharp slowdown in growth in the economy in the second half of 2001. The Central Statistics Office provisional estimates showed that the economy grew on an annual basis by 12% in the first quarter of 2001 but had slowed to zero by the fourth quarter of 2001. As a result, income tax receipts in the first six months of 2001 grew at a rate of almost 9% but fell by over 3% in the second half of 2001 compared to the year 2000. It would have been expected on that basis alone that income tax receipts in the first half of 2002 would have performed poorly in comparison to the first six months of 2001.

Looking forward to the second half of the year, the Department of Finance expected revenue to improve in response to a stronger economy. Nevertheless, it was expected that tax revenue for the year as a whole would be of the order of €500 million less than the Budget estimate, due for the most part to lower than expected economic growth and higher than expected SSIA costs given the unexpectedly large number of accounts opened as the "window" came to a close in April 2002.

In Depth Examination of the Income Tax Base

The Accounting Officer informed me that the Department, in consultation with Revenue, is examining the income tax base in more depth based on the up-dating of the income tax file to 1999-2000, supplemented by estimates based on latest PAYE returns to end-2001 and an examination of trends in PAYE payments by a sample of larger employers. That examination was ongoing. It was considered that the major changes made to the tax system in the 1999 and 2000 Budgets in relation to tax credits and widening of the standard rate band may have affected the income tax base and the projections that could be made on that basis. Furthermore, the elasticity of tax receipts at a time when GDP was increasing was likely to differ significantly from that experienced when economic growth slowed sharply, especially if overtime, bonuses or commissions were being cut as growth slows. Such income was more likely to be taxed at the higher rate. He stated that the issue was one of concern to the Department and was being carefully examined and monitored closely in preparation for the next Budget in December.

Mr. Purcell

For the benefit of committee members, we carried out a value for money report, two or three years ago, on the enforcement process in Revenue and looked at the various instruments and how they might be made more efficient. Traditionally, up to perhaps two years ago, I would have included a section in my annual report on enforcement activity and the results and the outturn from it. I have felt, and I am open to the committee's views on this, since it was being produced by Revenue anyway in its annual report, there was no point in replicating that in my annual report. If the committee thinks otherwise I can certainly go back to the former practice.

Section 4.8 deals with VAT repayments. VAT receipts in 2001 amounted to some €10.5 billion but over €2.5 billion was repaid to registered traders in that year. This section of my report records the results of an examination of the Revenue system for repaying that VAT. Repayment of VAT arises in any period when the VAT on a trader's purchases is greater than the VAT on sales. For some traders this can be an occasional occurrence, for example, when an expensive piece of equipment is purchased. However, for others it is a permanent feature, for example, in the food industry where VAT may be payable on certain purchases while sales are zero rated.

Much of the repayments system is automated. This enables the Revenue to issue payments to traders within ten working days of receipt of claim in most cases. The commitment of Revenue to customer service has to be balanced against the objective of only making correct repayments and the minimisation of the likelihood of fraud. Due to the nature and scale of the repayments it is potentially an area that is susceptible to attempted fraud. A few cases have been detected in recent years, including a major one involving a senior Revenue official which has occupied the minds of this committee in previous years.

The results of my examination suggest that, by and large, Revenue is close to getting the balance right between the competing demands of customer service and internal controls. The report recommends a number of areas where controls could be strengthened and Revenue has taken these on board. We heard one of them earlier regarding the rotation of staff. In the medium term, the early implementation of a computerised risk analysis system and its application to VAT repayments is the key to effective management of the control of risk in this area. I understand this is moving on apace.

Section 4.9 deals with dividend withholding tax. This is a fairly new tax so I thought it would be important to get in at an early stage to establish that there was an effective control framework in place, bearing in mind it is a tax that relies heavily on the companies and agents paying out dividends for its operation. My work built on a 2001 internal audit report on the administration of the tax. From its introduction in April 1999 up to the end of 2001, net amounts totalling €366 million have been collected through the medium of this tax. As its name suggests, it is a means of collecting income tax at the standard rate upfront from dividends paid to Irish residents by Irish-resident companies. In that respect it is somewhat similar to DIRT and that was part of the reason I went in at an early stage to ensure matters were on a sound basis. Recipients are required to state the amounts of dividends on their annual income tax returns and may be liable for tax at the marginal rate. Matching information from dividend withholding tax returns with income tax returns should help in validating the statements made by income taxpayers regarding dividends received.

It was clear at the time of the examination that initially Revenue's focus was on legislative, customer and IT development issues. To an extent this is understandable. While it had implemented basic controls over the scheme it had not instigated a systematic monitoring regime over compliance by companies. Some of the consequences of not doing so are referred to in the report, such as the failure to follow up on companies who had not filed a return for a current year even though they had filed returns for earlier years. I am now informed that liability totalling €193,000 has since been established on five companies as a result of that particular operation. That is not a whole lot in the context of the €366 million but in light of what happened on the DIRT side, we should be careful not to let matters slip.

I am glad to say that now that the building blocks are properly in place, greater emphasis and resources have been assigned to scheme compliance issues. The details are set out in the bullet points on page 50 so I will not repeat them. To summarise on section 4.9, things have been tightened up and further improvements are being implemented as we speak.

Section 4.10 resumes the discussion on the forecasting of tax receipts. The committee last considered this section of the report when it had the Accounting Officer of the Department of Finance before it on 23 January. At the end of that meeting it was decided to defer final consideration until the Accounting Officer for Revenue had been examined. The net issue was the €2.5 billion shortfall in the tax take in 2001 as against a forecast for that year and a continuation of that trend into 2002, albeit at a lower level, €1 billion. Although acknowledging that tax forecasting is not an exact science, the scale of the shortfalls suggested that the methodology employed needed to be looked at again.

Apart from the variables that stem from unexpected economic movements, there appeared to be a particular problem with income tax estimation, specifically on the PAYE side where there was a shortfall of over €700 million in 2001 and 2002. Members will note that in correspondence with the committee on 11 February, in response to a request by Deputy Higgins, the accounting officer stated that in an effort to address the problem Revenue has moved to improve the reliability of incomes data used to update the tax model for running budget costings by deriving a factual measure of income growth through the use of information drawn from a sample of more recent PAYE returns. This approach was first acted upon in 2002 in advance of budget 2003. Therefore, it is a little early to evaluate its effectiveness.

I will be brief. With regard to paragraph 4.8 on the repayments of value added tax to registered traders, the Comptroller and Auditor General has noted that VAT as a tax has unique administrative requirements in that one quarter of the gross collection is subsequently repaid to taxpayers. The scale of the repayment operation can be appreciated by the volume of repayments. In 2001 there were over 250,000 claims and 163,000 separate repayments. As pointed out in the audit, some of the concerns identified are already being addressed by Revenue and others are under consideration. Given the volume of repayments and the amount of money involved, €2.6 billion, it is reassuring to note that the audit did not find a single case of a VAT repayment having been made that was not properly due, but I am conscious of the other recommendations and we are taking them all on board.

Paragraph 4.9 is the other project audit dealing with dividend withholding tax. As the Comptroller and Auditor General said, this is a relatively new tax introduced in 1999. Like any new tax, there is a certain settling in period before all administrative issues can be addressed or identified. Our approach to this was to deal with a hierarchy of strategic issues prioritised on a tax at risk basis. On the compliance front the initial emphasis has been on PLCs and other large corporates which account for over 90% of the tax and where I am happy to note that all returns and payments are received on time. The Comptroller and Auditor General and our own internal audit branch have pointed out a number of issues which require action. These have been addressed or are currently being addressed by a focused compliance strategy which addresses all the points made.

With regard to paragraph 4.10 on the forecasting of receipts, the Comptroller and Auditor General mentioned the background to this. In my reply to the Comptroller and Auditor General, I described our role in the forecasting process as one of supporting the work of the Department of Finance. I outlined broadly the estimating methodology used and, while identifying known causes accounting for some of the shortfall in 2001, concluded that a major contributing factor was a slowdown in the rate of economic growth reflecting a general slowdown in consumption exacerbated by the foot and mouth crisis and influenced in the last quarter by the attacks on the US on 11 September. In commenting on the quality of our forecasts, I pointed out that forecasting tax receipts is not an exact science and is very much dependent on a number of key growth factors which are themselves forecasts. While every effort is made to have reliable basic data, there would also be scope for some margin of error. I also assured the Comptroller and Auditor General that we had examined the question of whether a reduction in the efficiency and effectiveness of collection contributed to the shortfalls and we were fully satisfied - it was important to me that I would be satisfied in this regard - from looking at certain compliance measures that there was no such reduction.

The Comptroller and Auditor General found that the Revenue Commissioners possessed very little management information regarding VAT repayments other than the number and the value of the repayments made. The Revenue Commissioners seem to be under utilising additional information it could receive from various return forms. This information ranged from trading details on return forms, which will assist in the risk analysis of claims for VAT, refunds, or repayments to the analysis of supporting financial documentation on corporation tax return forms to using the PPSN on a dividend withholding tax return form. Why have the Revenue Commissioners not systematically used all such information at their disposal?

It was difficult for us to match all the information we had in the past. We had a good airing of this matter at a previous meeting of the committee. The Chairman will know our view in Revenue is that the use of a PPS number by everybody involved in business or having anything to do with Revenue in any guise would be of enormous benefit to us in matching information. We are advancing that agenda as much as we possibly can. The other element to it is the risk analysis system about which I spoke, which will bring everything we know into a system, will profile cases and information and highlight the cases we need to look at. We will build into that the use of management information.

Returning to the earlier discussion on the random audit, it is probably out of this system or by way of an add on to it that we will get the type of management information the Comptroller and Auditor General and the Deputy talked about, which will enable us to match the resources to the risk much more in future.

I call Deputy Fleming.

I have a number of questions and wish to make one observation.

Does Deputy Fleming wish to give way to Deputy Dennehy? I have Deputy Fleming's name on the list.

The returns in this area are complex having regard to the sheer volume of numbers and the volume of money involved. What safety measures have been put in place to avoid a repetition of the unfortunate situation we discussed previously concerning a former official who received a jail sentence? In that context, the Comptroller and Auditor General mentioned where procedures can be overridden and a potential weakness where manual intervention allows for system checks to be overridden. Is Revenue satisfied that checks are in place to ensure that cannot be done again?

Generally speaking, as a consequence of this audit, we have reviewed our procedures and there are many recommendations in the audit. I could give the Deputy an update on each one, but that might prolong the proceedings too much. All I can say is that we have generally reviewed all our procedures. Rotation of staff and of the hierarchy of staff who are signing off on VAT repayments is part of the answer to the Deputy's question in regard to the circumstances in which a former official was dismissed by the Government last November. It brings that episode to a close. The committee asked about that previously.

I am concerned to ensure that the gap is plugged. I like the Revenue's target figure that 85% of claims will be repaid within ten working days, but the figure in respect of those who were not repaid over 30 days is still 6%. Did Revenue suffer any penalties under the 35 day clause in the prompt payments legislation? Is Revenue governed by that provision?

I do not believe it currently applies to VAT repayments. There is a provision this year's Finance Bill for the payment of interest on overpaid tax.

That legislation was drafted to guard against some people who might squirm out of their obligations. The period to which I referred is a long time for people to wait for a repayment.

Figures for 2002 show that we managed to make somewhat of an improvement in that 81% of people are now repaid within ten calendar days. The 30 day figure is still at 93%. The problem is that some of these cases tend to be very difficult and sometimes it is necessary to do a verification audit, as we mentioned earlier. With regard to the provision in this year's Finance Bill on the payment of interest on overpaid tax or delayed payments of tax, in so far as we needed to be galvanised, that will certainly push us forward. I must revert to the point made earlier by the Comptroller and Auditor General concerning the balance between speedy repayment of VAT claims and the need to have controls in place. By definition, controls slow things down.

I accept that but if the repayments of a small business with cashflow problems are outstanding for a month it can close such a company down in extreme cases. It is important to keep and eye on that.

What has helped there is that in excess of 99% of VAT repayments are made by direct credit - electronic funds transfer - into the bank account of the business. That has certainly speeded things up enormously.

It is surprising that the integrated taxation system has taken so long to come into effect, so that for all taxation purposes the taxpayer will be treated as an individual. In examining such matters we should worry about those outside the net. The Comptroller and Auditor General referred to phoenix-type operations with a poor Revenue history, involving traders registering with no fixed place of business. That is fairly topical at the moment. How does one get hold of these people, particularly for VAT, because they are literally cash businesses?

They are a special focus of our special inquiry branches. The special compliance districts, as I mentioned earlier, will be in the regions and will have a special focus on mobile traders - people who move quickly in and out of exhibitions and concerts. There is a provision in the VAT legislation for us to require such people to have a bond or deposit with us. We use that provision from time to time. We have tightened up on that area quite considerably. My IT people will kill me if I do not say that, while the ITP project has taken a long time, it has been one of the biggest such projects ever undertaken in any tax administration. It is at the stage now where almost all the taxes are in there.

It is not their fault but rather the fault of the legislators that there has been a total reluctance in this regard. We kept them separate for generations, even for once-off taxation, which was crazy.

The committee was concerned about capital gains tax and relevant contracts tax - the withholding tax - both of which are now fully in the ITP system and available for offset.

As regards the dividend withholding tax, Revenue has a captive audience as, I presume, they are all registered in the financial sector. There are 111 other companies, apparently. Can I take it that Revenue does not anticipate any problem with the dividend withholding tax? Perhaps I picked up the Comptroller and Auditor General's report incorrectly. Are they all within the system already?

They are but, in fairness, the Comptroller and Auditor General was making the point that at times we had perhaps not linked them. There were two main issues in this project audit. The first issue was that we did not have comprehensive procedures to ensure that all liable companies submitted returns. We now have a compliance regime for that. The other issue was that distributions as salary or other remuneration payments might not be properly returned for tax in the hands of the recipients. It is not a serious issue but we have taken action on various fronts as regards that. We have had an investigation and have passed it on to our anti-avoidance unit. I do not think it is a big problem. In terms of watching for problems that may emerge with this tax, however, the right approach is to try and anticipate problems, rather than anything else.

My final question refers to forecasting tax receipts. Bearing in mind what the Comptroller and Auditor General quite correctly pointed out, having regard to the serious implications for the economy of inaccurate Revenue forecasting, I find it almost too glib or easy to say that forecasting is not an exact science. Perhaps Mr. Daly could explain whether such forecasting is based purely on historical fact with an inflationary percentage built in. How does the Revenue go about making such forecasts? They are important, especially when there is a statutory requirement to assist the Minister for Finance. I take Mr. Daly's point about the two major issues that arose in 2001 but the figures were so far out that it is hard to accept the argument that forecasting is not an exact science.

In saying that, I did not intend to be in any way glib. The forecasting is based on a mix of items, including a trend analysis emerging from earlier years, and macros - including GDP, personal consumption and expenditure macros - which we receive from the Department of Finance. Macros in themselves are forecasts, so we have a forecast being built on something that is in itself a forecast. Historically, if one looks at GNP or GDP forecasts, which are made at the beginning of any year, or even through a year, and then one examines what emerges at the end of the year - as well as a definitive GDP for a particular year that emerges two or three years later - there is often quite an extensive variation. Due to the fact that the macros are so important they will impact on the forecast. We have tried to bring our own models and databases more up to date.

In the past, we tried to be too perfect in terms of the database we used. Our income tax database, for example, is based on the material we gleaned from the returns that come in from employers and the self-employed. In the past we would have waited until we had a very high percentage of those in before we did our modelling, which was a search for perfection. Of course, the result was that we were building our forecasts on material that was two or three years out of date. We have updated that process and are now building on a year that is much nearer the forecast year. The head of our forecasting branch, who is here, is much more expert on this matter than I am. I do not know whether I am answering the Deputy's question.

It is not even the historical issue of that particular year, it is a question of what changes will be applied because of the outturn. It affected every Government programme, although we have all worked on the promises made in some areas. I am interested to know what changes Revenue will make as a result of the outcome of that particular year.

We are talking about 2001, which was a freak year to a large extent.

I accept that.

It came after an extraordinary boom year in 2000.

Mr. Paddy Molloy

There are two issues here. The first is the issue of forecasting tax revenue on the basis of cashflow collection. We do that by taking the outturn for the year prior to the budget year, or an estimate if we are doing it before the year ends. We then take account of economically-based, year-on-year growth projections for the coming year, which we loosely call macros. These are macro-economic indicators of wage and salary growth, consumer expenditure, VRT from car purchases, etc. In that sense, we are starting to forecast from a pretty factual base, which is the outturn for the outgoing year. The extent to which our forecast, which is done initially on a pre-budget basis, is soundly based is heavily dependent on the reading of the economy for the coming year. As Mr. Daly said, looking back - one, two and three years on - to the original budget year, one can find quite significant changes - upwards in boom years and in less buoyant times the revisions would be significantly down compared to the original batch of forecasts.

The second issue is where we have to minus off the cost of budget concessions. That must be done before we finally have what we call a budget estimate for the year. We have a pre-budget version which takes no account of budget changes. The taxpayer model is used when considering budget changes. In a year of significant, expensive budget changes, the chances of under-measuring the cost of the changes increase with the greater the change to the system. The cost of the budget changes we first measured for 2001 were significantly out compared to what we would now say on a look back basis.

Our problem was partly due to the fact that our history model of taxpayers was three years older, or thereabouts, than the budget year. Given this, we must build on the differences, for what may be termed the gap period, by using the economic growth forecasts we get from the Department of Finance. This will bridge the gap in time. The revisions to the economic forecasts can cause us to revise our earlier budget estimates. One way to improve the reliability of the forecasting technique is to shorten the gap between the history year and the budget year by taking an early look at history data rather than relying on projected economic growth factors. We take a look one year later at what may be described as an early photograph of a set of incomes data one year beyond the normal history year on which we would rely, to eliminate the need for relying on forecasts of income growth.

It is a stepped, or building brick, operation. It might be easier to explain by way of example. In costing the budget changes for 2001, we were working from a history year of 1997-98. If we were doing that now we would have knowledge of 1998-99 as well. However, thereafter, we would not have knowledge from the tax file because we would not have a sufficient quantity of returns to give us information on PAYE, wage and salary earner income levels. It is a returns driven process.

In the final analysis, the forecast of tax revenue has to depend on a reading of the economy generating a forecast of economic growth. If unexpected things happen during the year which have a strong impact, such as the foot and mouth disease, the world economy effects of the attacks on the United States or an ongoing recession where some lift might have been expected during the year but did not happen, the forecast at budget time compared with what we might see even six months later can be quite different. There is a risk in a yo-yo situation, where one is not on a clear downward or upward curve and where hills and valleys can emerge during the year and throw forecasts out of kilter.

I understood it was based on a historical perspective. For example, the committee ceased to examine reports that were four or five years out of date because it was a waste of time. I fear the same thing might happen in this instance. Virtually every day, business programmes produce new forecasts on the macro side for the coming year. Given the speed of events, it is frightening to rely on what happened two or three years ago. We should give up that process.

There is a need to change the procedures, which the Minister for Finance might welcome. Given your experience, will the glitch that arose in the year under review change the methodology and bring the historical data more up to date? Will you try to deal with more immediate facts and figures rather than the historical record?

Mr. Molloy

The model and the two year aspect only dictate our measure of the cost of budget changes to the system. The budget change is only a fraction of the forecast yield. The main forecast of yield, the money, is based on the hard cash we collected in the year before the budget. The main obstacle or factor in deciding the correctness of our revenue forecast for the coming or budget year is the reading of economic growth. Budget changes can be seriously under-costed if they are large in scope or if we have not taken proper account of income growth over the period where the model lies into the past. In recent years, in 2002 and much less in 2003, the impact of the budget changes are much less in terms of the proportion of the total tax yield. The less they represent the less we must worry about erroneous calls.

I thank you for that fascinating explanation. We could perhaps consider this matter more fully at another forum. I have concerns about how some figures have been arrived at. Everybody was being blamed, including the CSO, the Department of Finance and the Revenue. It was a freak year, but we have much to learn from it.

Before calling Deputy Fleming, I will take a comment from the Department of Finance.

Mr. Dermot Mulligan

In case of a misapprehension, I assure the Deputy of the importance of economic forecasts about the future in terms of the tax forecasts. While my colleagues in the Revenue Commissioners have explained how they use historical data as an input into the development of their tax forecasts, a significant factor in making forecasts on future tax yields is a consideration of future economic growth and its likely pattern. It is not simply a question of assuming that past economic growth patterns will continue. We try to make as good an estimate as we can on the rate of future economic growth, what kind of growth it will be and how it will impact on the different tax heads referred to by my colleagues from Revenue.

The Deputy asked why the tax forecast turned out to be so wrong. The economy played a very big part in that. In 2001 it would have been difficult to predict the onset of the foot and mouth disease outbreak, the scale of the slump in the IT sector and the events of 11 September in the United States. These all played a significant role with regard to the development of the economy in 2001 and 2002, when most commentators expected a world-wide economic recovery which would help the Irish economy. That did not happen as expected.

When we framed the last budget we were aware of the possibility of war in Iraq, which would have been included in our economic forecast, including the likely impact on tax receipts. I do not want the Deputy to be under the impression that we simply roll forward whatever has happened in the past. We do our best to predict the likely rate and pattern of economic growth.

We should take comfort for what happened in 2001 because of the obvious explanations. The forecasts for the previous three years were also out, but thank God the out-turn was so benign. Nobody foresaw the boom. I am concerned about the methodology because it affects us all. On this occasion we are unfortunately getting the hammer because anticipated financial resources are not available for schools and hospitals. I am sure we will learn from the mistakes inherent in the methodology.

Up to 111 companies account for about 92% of the dividend withholding tax paid. The chart on page 46 of Mr. Purcell's report states that in 2001 about €143 million was collected. That represents the deduction at the 20% standard rate which was withheld by the people giving the dividends. I want to concentrate on the balance of moneys which might be due at the higher rate of tax. How many taxpayers declared on their tax returns that they had received a dividend, which had been the subject of withholding tax and which would lead to either a balancing amount being paid or, if the taxpayer was at the standard rate, no additional tax being paid? Certainly in quite a number of instances, there would have been a balancing amount to be paid.

The reason I am conscious of this is that page 21 of the Revenue's annual report, which refers to the key statistic for dividend withholding tax, mentions 2.3 million such distributions. We all know that most of the companies probably make two distributions in the course of the year and, therefore, one can divide that figure by two. I am thinking of the big companies where there are 500,000 shareholders like Eircom. I am also thinking of Permanent TSB and other companies which have recently gone public. Each year there must be between 600,000 and 800,000 taxpayers who each year are receiving dividends, albeit for small amounts. On average, most of that number must be paying tax at the higher rate because they would be in a better position to afford to buy shares than those on the lower rate.

Given that €143 million was paid at the 20% rate, how much extra was collected from people making the balancing payment at the higher rate? How many people declared such amounts on their tax returns and was it in that order?

I noted that the Comptroller and Auditor General's report refers to a sample of 50 recipients. Although that is interesting, the dividend income of €1.6 million would mean, on average, that each case was in receipt of dividends of €32,000.

What is the question?

Therefore, they are purely the top 1% or 2%. They are not representative of the 600,000 people to whom I refer. Can the Revenue give us some information on that and how it is proposed to be dealt with?

I do not have the type of figures the Deputy is looking for. One of the elements of our compliance programme is to make sure that the balance of tax is collected, that taxpayers declaring dividend income are identifiable. We now have a compliance programme in place which, in effect, means that we are picking up from the dividend withholding tax returns the payment of dividends to particular individuals. As part of the compliance programme, that information is being fed into tax districts. Therefore they will begin to match it with ordinary tax returns to make sure that the income is being returned and that the balance of tax at the higher rate is actually being paid.

I am not sure I have the figures sought by the Deputy but I can certainly see if we can get them. I am not sure that we will be able to get them. I can assure the Deputy that there is a compliance programme in place to make sure that we can pull from the dividend withholding tax returns information on individual taxpayers and we can match that with their ordinary taxpayer returns.

There is a slight difficulty in this in that we do not, at this stage, require the PPS numbers of people to be shown in dividend withholding tax returns. That was considered at the time, but it was decided at that stage that it would be quite a significant burden on companies and the type of companies the Deputy mentioned would have thousands of shareholders. However, that issue has not gone off the table because it goes back to the point I mentioned earlier about the PPS number as being the necessary link for Revenue in all of this.

I look forward to receiving whatever level of information the Revenue has on that. I realise that the perfect solution will be to have the PPS number on all dividend cheque payments. Notwithstanding the Revenue's publicity campaign and other campaigns, I suspect there are hundreds of thousands of people who, perhaps unwittingly because of the small amounts involved, are not declaring their dividends. They forget that it is not part of their normal source of income and is an exceptional item. I hope the numbers of people who declare such a return will be in those high figures.

There is now a specific question on the return form about dividends paid.

That is the information I am looking for. I am sure the Revenue has those figures on the computer system.

We will see if we can get that.

I want to refer back to the forecasting of tax receipts. Does Revenue or the Department of Finance have the estimated figures for income tax receipts for the first quarter of the current year?

We would have the profile figures and the outturn. Is it income tax, in particular?

Income tax and VAT, just two items.

I will give the estimate first, the profile as on our website. The projected figure to the end of March for income tax, including PAYE and non-PAYE, was €2,164 million and the figure for Exchequer receipts is €2,011 million. The profile for VAT was €2,790 million and the outturn is €3,010 million. Therefore, there has been an increase in VAT and a shortfall in income tax.

I accept fully that 2001 was a freak year and there were two major incidents, but I want to discuss the methodology and procedure because this is quite important. The last page of the report, page 57, contains a paragraph on in-depth examination of the income tax base. I was a little taken aback when I read the following towards the end of the paragraph:

Furthermore, the elasticity of tax receipts at a time when GDP was increasing was likely to differ significantly from that experienced when economic growth slowed sharply, especially if overtime, bonuses or commissions were being cut as growth slows. Such income was more likely to be taxed at the higher rate. He stated that the issue was one of concern to the Department and was being carefully examined and monitored closely in preparation for the next Budget in December.

If that aspect had not been in the previous methodology which the Revenue was using, to use this elasticity as an example from the report, I would have concerns about the reliability of the methodology we are using for this forecasting. I am surprised at the outcome figures for the two examples: income tax is probably 6% or 7% lower than expected, and VAT is 10% higher in the first quarter. They are significant swings. I would have hoped that forecasting methodology was tighter than that and used figures that were much closer in time.

Mr. Daly mentioned that the impact of the budget changes was now less significant. I know there was a change in the VAT rate, but that was accounted for. I am surprised that the margins for the current term, which is the most recent, are still quite large - 5% or 6% down in income tax and 10% up in VAT. I question the overall methodology. In particular, having read that last piece about elasticity, I interpreted it, rightly or wrongly, as meaning that the issue of elasticity had not been accounted for in previous terms. If so, that is worrying. My interpretation was that this elasticity in the income tax base would only be accounted for in that estimate.

In any event, I am still disappointed with the margin of outturn versus estimate and I am concerned about the methodology. The consequences for the future we all know, but 10% on VAT is a very significant margin.

I will make a very quick comment on the 2003 figures and my colleagues from the Department of Finance may wish to comment on the elasticity question. I caution people not to read too much into the first quarter figures of any year, particularly when it comes to income tax. Most income tax is now payable towards the end of the year in October. What the members may be seeing here is the impact of the pay and file campaign we introduced last October, which made sure that payments on account and so on were very much up to date. We had much less of the residual fallout that we would usually have in the following months. On VAT, in January to March we were seeing the pre-Christmas sales effect. Perhaps this is something we need to profile better throughout the year, but I caution people not to read too much into the first three months.

I do not buy the answer about VAT. I asked Mr. Daly specifically about the first quarter. People know when they are forecasting for the first quarter that December shopping and January sales will have to be taken into account. It is precisely those things that forecasting should be taking into account. They are the details.

That is the point I made: maybe our profiling, month by month, needs to be better. We have discussed building on the historical base. I am very anxious to promote getting away from the sometimes sterile analysis of GNP and GDP and getting closer to business. What do businesses think about how business activity will go during the year? We can get close in some areas. We have a much closer relationship, for example, with people in the excise area, in which there are a relatively small number of people in the alcohol, tobacco and oil sector as well as the vehicle sector. They are usually the best people to tell us what their business will be like during the year. We are getting closer and closer to those. Corporation tax is more difficult because companies are obviously not going to tell you what their profits will be - they would have problems with the stock exchange and so on if they did. We have very close contact with the IDA, Enterprise Ireland and so on to see what their forecast is for the year.

I accept the Deputy's point, however. Overall tax revenue for the first quarter is broadly on target. Between tax heads it varies.

No, Mr. Daly, that is a point that would cloud the issue. What I am talking about is forecasting and its methodology. It is more random luck that the outturn of a bundle happens to match.

I accept that. I am not arguing about that.

It does cloud the issue I am specifically trying to address, namely the methodology of forecasting. Mr. Daly is right - the outcome for the whole bundle of taxes happens to be close enough, but that was pot luck.

We are agreed that the profiling needs to be improved, but it is a very difficult process.

I am not disputing that.

Mr. Mulligan

The paragraph on page 57 seems to be saying that the elasticity of tax receipts and their performance compared to that of GDP moved in a particular way and the reference to bonuses and overtime for income tax was given as an explanation of the way it moved. The paragraph is about the overall performance of tax receipts compared to growth, whether measured in GDP terms or GNP terms. It is not an income tax elasticity that is referred to but total taxes compared to GDP. The comment on the way the elasticity has moved in relation to bonuses and overtime is intended as an explanation of that movement. I do not know whether that helps.

Referring to the first quarter figures, I would echo what the chairman of the Revenue Commissioners has said. The first quarter is only the first three months of a year that is yet to develop. In terms of the first quarter, the Exchequer position is more or less where we would expect it to be, given our targets for the year as a whole. I take the Deputy's points in relation to individual subheads and how they have developed. Some subheads have done better than expected in the Revenue Commissioners' profile and some have not done so well. For example, many commentators have said that the unexpected strength of the housing market has been a factor and there are other explanations for the way others have developed. Whether those factors are sustained over the rest of the year remains to be seen.

I know these are explanations, but I am still concerned about the methodology of forecasting. I know that the total revenue is more or less on target and I appreciate that, but I still think an outturn for VAT that is 10% or more off target is worrying in terms of the methodology used to produce those figures. The error of 10% works out at €200 million or €300 million and when this is multiplied out over the course of the year, the margin of error is quite significant in terms of the total volume of money. Some 10% does not sound that much, but the trend continues. It is the methodology I am concerned about.

I agree with Deputy Curran's point that so much depends on the accuracy of the figure. Departmental commitments, particularly in the health and education sectors, depend so much on the accuracy of these figures. A lot of assumptions can be made but the projections may not match up. From a business point of view, as Deputy Dennehy said, when the trends are factored in the error should be considerably smaller. Deputy Curran has indicated his concerns clearly. It is quite difficult to get a defined figure, but with the current infrastructure deficit, when the figures are out by so much the impact can be very wide.

Regarding VAT, can Mr. Daly tell me the extent to which administrators' costs are being eaten up by the low threshold for the registration of payment of tax for sole traders, for example? It is a very low threshold. Is it €25,000 or €50,000? It is certainly not much more than an individual wage. I imagine many of the repayments would have to be made to this sector. If the threshold were raised, it might help Mr. Daly and the type of work he does in this area.

I would not be sure that many of the repayments would be going to the lower sector. I do not have the type of figures as to what the resource cost for Revenue is in relation to this. I know the threshold is looked at from time to time and we balance the burden on the taxpayer and on Revenue against the compliance value of having the threshold at that level. It is at €51,000 where goods are supplied and €25,500 in the services industry. It is not something that would be in my mind right now. We would have a view that having people registered for VAT is a very good compliance tool. I would not have a figure as to what it costs us to police or control that threshold or what it would cost if we increased it by 10%, 15%, 20% or whatever.

My question on the dividend withholding tax is again a bit left-field. Is there a way for Revenue to measure the extent of tax avoidance by companies and their shareholders where this tax is concerned? There would be an incentive not to pay dividends and for people to pay a higher rate of individual tax if the money were retained within the company to pay a lower rate of corporation tax. It might be given to shareholders by way of increased shares within the individual companies as an avoidance mechanism.

We have no way of measuring it but that very point is being considered by our anti-avoidance unit and we have increased the staff in that unit in general terms. That is one of the items on their agenda and is something of which we are conscious. I am not aware that there is a big problem but we have come across one or two cases that lead us to think that it is something we should look at.

There should be an obvious correlation between dividends and corporation tax paid out in a given year.

That is the sort of study that will be done in this area.

Perhaps this would be better directed at the Department of Finance. In terms of forecasting, I would have thought we would have learnt by now that gross domestic product, GDP, is a notorious tool to use in the forecasting mechanism, especially for the Irish economy. The variation between GDP and gross national product, GNP, is enormous because of the degree of foreign direct investment, the level of repatriated profits and the practice of transfer pricing. Is ignoring GDP in the forecasting mechanism the best way of ensuring that figures are more balanced in future?

I think that is a matter for my colleague.

Mr. Mulligan

The Deputy is right. Certainly there has been a significant divergence between GNP and GDP recently. Whatever methodology was published by the Department back in 1998, there was not quite the divergence at that stage that has happened since. As a result of that, we also look at GNP developments whenever we look at tax forecasting.

The Deputy should be assured that it is not simply a question of mechanistically implementing a methodology. It is a matter, as I think it was described before in the committee, of using it as a rule of thumb to see how things are developing between the change in economic growth and the change in the performance of tax receipts and whether the elasticity calculated gives rise to any questions that need to be looked at in more detail.

The Deputy should be assured we are not simply mechanistically looking at GDP and that we look at GNP. It is an extremely important indicator in relation to the changes in the economy and also the implications for tax receipts and tax forecasts going forward.

The problem is that it is using a thumb after it has been hit by a hammer. That is bringing about much of the exaggeration in the receipts.

What are the broader implications of the High Court decision in November which ordered Revenue to pay interest on VAT repayments to the Bank of Ireland? Will that have an impact? Was much paid back?

I do not think anything has been paid back as of yet, Chairman. I am not sure if it is under appeal. Discussions are ongoing with the agents of the Bank of Ireland as to what the quantum will be. The more general problem of payment of interest on overpayment of tax has been dealt with in the Finance Act.

Has it been agreed that you will pay it back?

No, I think discussions are still ongoing.

It was said at the time that this would result in a flood of similar claims. What would your opinion be on that?

I honestly do not know enough about it and I might be misleading you, Chairman, by giving you a definitive answer. All I know is that we have had the High Court decision. Discussions are still ongoing with the agents for Bank of Ireland in relation to the quantum, detail and what might be the outcome. In advance of that, we have not looked at any other companies, but in so far as other companies are concerned, the circumstances would have to be the same. I would be misleading you if I gave any definitive answer.

You might come back to us on it.

I will brief you.

On the issue of VAT and the €634 million paid back to companies which had not been audited since 1990 and new companies, what procedures are in place to monitor that type of repayment by Revenue?

We would have made the point in response to the Comptroller and Auditor General's report that, maybe because of records of audits not being put on the ITP system, it could not be relied on as a record of a company not having been audited since 1990 or whatever. As well as that, a lot of repayments would have been to companies in a group situation where the group might have been audited and it might not have transferred back.

I do not think the picture is as bad as it appears. It is simply that our systems did not record audits. Again I go back to the fact that, by and large, every company comes within a screening process at least every three or four years.

The other point made by the Comptroller and Auditor General in that part of the report was that sometimes we repaid VAT to companies that had only been set up for a short time. Our processes for registering companies for VAT are now much more rigorous than they were in the past and we inquire much more thoroughly into the company, business, directors and the people behind it.

Have you had the experience of a trader making a second return for the same period after the original VAT repayment claim has been processed? Has that happened?

That can happen and quite legitimately. In some cases people are in a net repayment situation and are allowed to put in more than one return in a period. The general point made in the report was that we should have a mechanism whereby there is a recording of these second claims to ensure there is no question of a duplicate payment or anything like that. That is one of the things that has been done and that system is now in place.

Would you agree that companies that have not been audited are clearly a higher risk than those that have and that those which have not been audited prior to 1990 or are new companies would be a higher risk than companies which would have a track record of payment and repayment for the refund of VAT?

By and large, yes, but sometimes the reason a company is not audited for a particular number of years is because we have done a risk assessment and have found that it is not in a risk area. I take the Chairman's point on that.

A dividend withholding tax avoidance scheme was uncovered which involved the payment of salary and other benefits being made by companies in the form of distributions that would initially attract the standard rate of deductions and avoid PRSI payments. What action have the Revenue Commissioners taken to close this loophole? Has the anti-avoidance section an estimate of how much revenue was involved in this scheme?

The point I made in response, I think, to Deputy Fleming was that we have no evidence that there is widespread use of that scheme but it is something about which we are a bit concerned. It is now on the agenda of our anti-avoidance unit to have a look at that and see whether there is that type of activity. I can certainly at some stage.

Do you see any integration of the DWT system and income tax in the future?

The real key here is that the companies making DWT returns to us should, at some stage in the future, have to show the PPS number of the recipient of the dividend. That will automatically enable us to link the dividend to the tax return——

That integration would be effective.

It would be very effective and goes back to a discussion we had at the last meeting about universal use of the PPS number, which Revenue is very keen on.

Have the Revenue Commissioners uncovered any other anti-avoidance schemes?

Not that I am aware of. We close off anti-avoidance schemes from time to time. Sometimes we discover things and make recommendations to the Minister but not in that general area of the DWT.

What is the normal procedure with loopholes or tax avoidance in the tax system uncovered by Revenue? Are recommendations made to the Minister?

There are a couple of ways of doing it. If it is something that needs to be closed off very quickly, as happened in some cases recently, we would go to the Department or the Minister for Finance and tell them it needs to be closed off quickly. The Minister may do so by making an announcement and then legislate for it subsequently in the Finance Bill. Making an announcement on a particular day means the measure can be retrospective and the scheme is closed off. There are other schemes we would deal with in the Finance Bill every year if there is a looseness in the law. My colleague reminds me we also challenge schemes in court, which we have done quite successfully.

We are talking about forecasting the potential revenue take and the loopholes in the tax code, which can be operated successfully by tax specialists. It is important now there is less money coming in that there are few such loopholes.

I agree and we can see in the last year or so there has been much activity in closing off loopholes and abuses. There are schemes and incentives built into the tax system for good reasons. People get allowances for particular developments, such as the IFSC, and there are sound economic reasons for doing so which are valid in the tax system. However, when a small number of people, for no valid economic reason but for greed, abuse these loopholes then they must be closed down. That has been done this year.

In cases where there is no substantive reason to avoid tax, are there people in the Revenue who check these schemes and incentives?

We review schemes from time to time and the anti-avoidance unit is always alert to this kind of activity. The inspector on the ground looks at tax returns and keeps in touch with activity in particular sectors; they would be very conscious of that. In terms of a restructured Revenue, there is a unit known as the large cases division which is focused on high net worth individuals and large corporates, which is where much of this avoidance develops. One of the key tasks of this new division is to get close to such activity.

I am delighted because this is all about accountability. The fact that some people can employ the best accountants to avoid paying tax means others, at the bottom of the pile, are deprived of essential services. I am delighted to hear about this.

I agree absolutely. In simple terms, the broader one spreads the load the fairer it is.

I am delighted Revenue is considering avoidance schemes and compliance at the highest levels of business.

If there are no other questions I propose we dispose of these chapters and note the Vote. Is that agreed? Agreed. I thank Mr. Purcell and his team for their excellent backup, as always, and I thank Mr. Daly and the delegation from Revenue. It was a very worthwhile debate.

The agenda for next week will be Vote 10 - Office of Public Works and Vote 44 - flood relief, as well as the report on the Office of Public Works's building maintenance service.

The witnesses withdrew.

The committee adjourned at 2.25 p.m. until 11 a.m. on Wednesday, 16 April 2003.
Top
Share