Could an amnesty be confined to funds abroad?
Such a "ring fence", despite its superficial attractions, would be impossible to operate in practice irrespective of the option chosen. Money within the country - in disguised accounts and bogus non-resident accounts with financial institutions - could readily be transferred abroad to avail of the tax remission. The more attractive the amnesty, the more likely this would be.
Should one try, nevertheless, to confine it to funds abroad?
Possible mechanisms to that end might include:
Require proof of conversion from foreign currency (but this would result in conversion of Irish pounds into foreign currency);
Some proof that the money had been abroad for some time (but it is not clear what form of credible proof could be sought that would not be open to manipulation).
In any case, confining the amnesty to those who had, in the past, moved funds out of Ireland, would be seen to penalise those who had kept their funds at home. Also, it would give an unequivocal signal that proceeds of future evasion should be put abroad. Not being able to confine it would, of course, mean another generally available amnesty on the lines of 1988, but, of course, on much more generous terms.
The Minister believes it would not be possible to confine the scheme to funds held abroad, and that it would be seen to be futile to attempt to do so.
What hold would we have on the repatriated funds?
In the absence of exchange controls, it is not clear how funds could be obliged to stay in Ireland. The suggestion has been made that there be an obligation to leave such funds in special investment media such as bonds, but this is inconsistent with the argument that the most likely clients of the amnesty are those having need of funds for personal business purposes, and has been contradicted in the press by some proponents of an amnesty. In any case, these media could be used as collateral for borrowings, whose funds could readily be expatriated.
What is the likely impact on the Exchequer?
The first question to be asked is what current revenue collection would be put at risk? There is a very significant yield from the various audit, arrears collection and enforcement measures operated by Revenue - 1989: £245 million; 1990: £207 million; 1991: £302 million; 1992: £322 million (estimated), and it is expected to be at least of the order of £320 million again this year. Depending on its precise terms, a tax amnesty in 1993 could cover a significant element of this amount with major adverse impact on the Exchequer.
Appendix 1 details the £320 million collected in 1992 and indicates the degree of risk to the various components arising from the amnesty. It shows that some £265 million is regarded to be at "high risk".
If 20 per cent of the on-going audit/enforcement programme were compromised, the loss would be over £60 million; a loss of 50 per cent would amount to £160 million. These annual losses would, of course, recur over a number of years given that audit/enforcement regularly refers back over a considerable period of years.
It could also be expected that there would be a definite negative impact on compliance generally, both in terms of the reaction, especially of reluctantly-compliant taxpayers, to a perceived inequity and of having grounds for anticipating repeat amnesties every four or five years. The progress made in recent years in making taxpayers more compliant and timely in their payments would be certain to be dented. As explained in Appendix 2, a slippage of even 1 per cent on normal current tax revenue receipts would imply a loss in 1993 in the range £64 million to £90 million.
On the other side of the scales, how much would the amnesty yield? There have been a wide range of estimates of the available pool of "hot money", and any assessment of how much might be attracted in is subject to wide variation depending on the precise terms on offer. If £500 million were disclosed - a round figure that has featured in media speculation - then the yield, assuming a 15 per cent "levy", would be £75 million. See Note 1.
This falls well short of the lower end of the range of prospective losses outlined above, even if the recurrence of these losses were disregarded. Furthermore, any attempt to limit that loss by tightening the terms of the amnesty or limiting its scope, would have the effect of reducing its attraction to evaders.