Value for Money Examination  21;   Value Added Tax Collection and Control   Summary

Value Added Tax (VAT) collection and control procedures are based on a self assessment system. The overall objective of the system is to maximise the collection of VAT revenue by maximising the levels of voluntary compliance and by deterring evasion. The principal compliance elements in the system are the checking of submitted VAT returns and the follow-up of overdue returns and payments (collection work), the checking of the accuracy of VAT returns through a VAT audit (control work) and the referral of debts for enforcement action (enforcement work).

There are two approaches to VAT collection and control work: an equity approach which applies the same collection and control procedures to all traders regardless of their size or VAT liability and an efficiency approach which schedules audit and enforcement work based on the amounts most at risk. In recent years the Office of the Revenue Commissioners (Revenue) has moved towards the efficiency approach.

The purpose of the examination was to review the current organisation of VAT collection and control procedures with particular reference to the results achieved from registration, collection, control and enforcement work over the past five years. The examination also sought to identify opportunities for further system refinements which would contribute to increasing the effectiveness of the procedures.

Registration Practices

Procedures performed at registration contribute to the effectiveness of the later collection and control procedures. The number of registered traders has increased steadily over recent years and stood at almost 150,000 traders in May 1997. Revenue relies on a mix of formal and informal information sources to detect unregistered traders who exceed the registration threshold. The examination found that the use of information obtained from audit activity and from the Companies Registration Office should be further developed to assist in detecting unregistered traders.

The detection of unregistered liable traders is hindered by a lack of knowledge of the extent of this area within the black economy. Several methods for estimating the population of these unregistered traders have been proposed but no reliable approach has emerged and Revenue has not attempted this exercise. The activity to detect unregistered traders is co-ordinated across all tax heads and is controlled by setting national detection targets. As the quality of information on detection rates improves, better estimates of the unregistered trader population should be possible which, in turn, should impact on Revenue's approach to detection.

At registration, traders are assigned a low, medium or high risk rating which is intended to be used for planning VAT audits but is not yet fully developed. The refinement of the risk rating depends on the quality of information available to Revenue about the trader's business affairs. Since 1995, there has been a policy for district tax officials to make new business education visits to traders to provide them with information about the operation of the VAT regulations and their obligations. These visits could also be used to gather information needed for risk rating. The examination found that targets for making education visits are not being met by some district tax offices and there is a need for Revenue to re-emphasise the importance of the visits.

Self Assessment and Collection

Revenue regards traders as being compliant if the VAT3 return and any associated payment is received in the month in which it is due. The examination found that there has been a gradual improvement in collection rates since 1993. By the end of 1996, on average, 69% of returns and 930/o of VAT revenue were collected within two months of the due month.

Where traders are not compliant, the Collector-General's Office may issue a Section 22 estimate of liability. In the past, there was a tendency to overstate the estimate to encourage the trader to discharge it by submitting a VAT3 return. Inspectors of Taxes may also issue Section 23 assessments which may adjust a VAT3 return or replace a Section 22 estimate. There was also an historical tendency for Section 23 assessments to overstate the trader's liability. As a result, for all VAT years up to 1993, there has been a build up of notional VAT arrears for Section 22 estimates and Section 23 assessments of ?62 million and ?105 million respectively.

Arrears Verification Units have been established in each district tax office to determine the true outstanding liability and reduce the level of arrears by discharging the estimates, collecting the amounts due or referring arrears for enforcement or write-off. The examination found that the Arrears Verification Units have made progress in reducing the VAT arrears. Furthermore, the practice of issuing Section 22 estimates and Section 23 assessments has been refined since 1994. This, together with improvements in the computer information systems, has increased the quality of information available to Revenue concerning the collectibility of the arrears.

An annual accounting scheme, where only one return and payment was required per year, was introduced in 1989 for small traders whose annual VAT liability was less than ?5,000. There was a high level of defaulting traders in this group and it was not possible to apply the normal collection procedures. While a special follow-up campaign has brought some results, Revenue now encourages small traders to set up direct debit arrangement5? which is a more effective system for collecting VAT.

VAT Audits

A programme of audits is performed to maximise VAT revenue by detecting underpaid VAT and to provide assurance as to the overall level of voluntary compliance.

Audits can be performed for single tax heads such as VAT or for all tax heads (comprehensive audits). The detection of underpaid VAT is best served by an audit programme which targets the groups where tax is considered to be most at risk. The assurance of the overall level of voluntary compliance is achieved by selecting some audit targets at random. Trade and industry organisations believe that there is too much audit emphasis on the collection of additional VAT and not enough on the assurance objective. The level of random audits performed is very low. No single tax head VAT audits are selected at random and only 52 out of 3,969 comprehensive audits performed in 1996 were selected at random.

Although it is inevitable that some audits will not result in additional VAT revenue (i.e. a nil yield), there is a higher risk that randomly selected audits will result in a nil yield. The incidence of nil yielding audits could be used as a performance measure for the appropriateness of the mix of targeted and random audits.

The planning of single tax head VAT audits has improved in recent years mainly due to the availability of better information and to the adoption of a sectoral approach. These audits are interrelated with audits under other tax heads. However, no formal risk analysis techniques or software are used in the planning process. The number of single tax head VAT audits performed has fallen from 13,158 in 1993 to 8,424 in 1996 due to a reallocation of resources towards comprehensive audits and an increase in the duration of audit to enable more VAT periods to be covered. The result is that the average yield per audit has increased from ?2,356 in 1993 to ?3,917 in 1996.

Although Revenue's accounting systems are not designed to record the full cost of performing an audit, the difference between the audit yield and the direct manpower cost suggests that VAT audit activity is good value for money and that the allocation of more resources to audit would further increase the yield. Revenue should improve procedures to monitor more accurately the cost effectiveness of VAT audits.


Enforcement procedures are not confined to particular tax heads. The principal methods used are sheriff and solicitor enforcement. Certificates covering specific tax arrears are sent to sheriffs for action in situations where it is believed that the trader possesses business assets which may be seized and used to recover the tax debt. Solicitors, including the Revenue Solicitor, pursue cases considered unsuitable for sheriff enforcement.

The costs and poundage involved in sheriff enforcement are borne by the defaulter. This acts as an incentive for settling liabilities directly with the Collector-General's Office. Approximately ?343 million (25%) of the face value of certificates issued in respect of VAT arrears in the period 1993 to 1996 was collected directly or indirectly as a result of sheriff enforcement. The examination found that sheriff enforcement has a higher relative success rate than the use of Section 22 estimates or the issuance of reminders and demands.

Cases amounting to ?60 million in VAT arrears were referred to solicitors for enforcement between 1993 and 1996. A total of ?18 million was collected during this period.