Dishonest conduct by tax agents

In follow up to its much criticised proposals in 2010, HMRC has had another go at how to deal with disreputable tax agents. The closing date for comments in response to the latest proposals is 16 September.

In outline, the latest proposal is that HMRC can accuse an agent (including an employed individual), acting in the course of a business, of dishonest conduct. The individual has the right of appeal to the FTT. If the accusation is upheld (the burden of proof being reversed, with it being for HMRC to satisfy the tribunal), the individual can be penalised up to £50,000 and "named and shamed". Thereafter, but only thereafter, HMRC can seek a notice from the tribunal permitting access to the agent's working papers. 

It is difficult to know what to make of the latest proposals. We should all like to see bent agents neutered, and if only to create a level commercial playing field. It is, further, welcomed that, contrary to earlier proposals, HMRC should only be able to seek access to working papers once it has been held by an independent tribunal that the agent is, indeed, cuplable. 

Notwithstanding, it is hard to see, firstly, why a civil approach is merited nor, secondly, why a penalty of only up to £50,000 is considered a satisfactory deterrent or punishment. 

It might be that the civil approach is favoured because it will enable employees to obtain, or demand from their employers, insurance to cover legal costs. But if so, that would be sufficiently important that it needs to be expressly stated as being part of HMRC's thinking.

HMRC makes only a scant attempt to address its very obvious desire to make the issue a civil one. The justification seems to be cost-based which is curious when, elsewhere, HMRC is more than happy, with the CPS, to spend disproprtionate sums in pursuing prosecutions for "low-level" tax credit cheats who are never going to represent "a centre of infection" in the same way as a crooked tax agent. 

It, further, has to be remarked that the test of dishonesty is very much the preserve of the criminal law, and is not one that HMRC is known to be good at understanding; and, indeed, why should it, when the functions of HMRC are predominantly civil in nature.

All told, for HMRC to have the first call in applying a test of dishonesty and, then, only having to evidence that conduct to the civil standard of the balance of probabilities, is unlikely to reassure many of HMRC's motives.

The proposals will be broadly welcomed, but only because they are better than the previous version. But that does not make the new proposals worthwhile. Nor, despite it having been raised previously, does it satisfactorily explain why there is a need for the existing laws have to be supplemented with any new ones whatsoever.

HMRC have said, and will continue to say, that any new law will be applied only in genuine cases of disreputable conduct. If so, the deserving cases will not be that many in number and the extant criminal offences under the Fraud Act 2006 or the common law of cheating the public purse would not only seem more than adequate, but would, further, allow for the Proceeds of Crime Act 2002 to be used for asset confiscation. 

In conclusion, unless HMRC can come up with a far more credible and fullsome explanation of its desire to devise "an effective civil approach", it is difficult to be convinced that HMRC has a case. Few of us were left in any doubt by the original proposals that HMRC were most interested in obtaining agent's working papers. With HMRC's enthusiasm for that having now been shouted down, the dropping of a new civil offence in favour of the existing (and, in many instances, fearsomely draconian) criminal laws would seem preferable.

Perhaps, all that is really needed, is for HMRC to identify a handful of crooked agents, and have the CFG bring some well chosen criminal prosecutions. Throw in some draconian confiscation orders, and a minor spend on advertising which probably would not even cover the cost of the average Dave Hartnett jolly, and it will be job done.