As the FT noted yesterday:

HM Revenue & Customs omitted key figures from its business case about how much its Making Tax Digital programme will cost the public, a report by the National Audit Office has revealed.

A report released by the NAO on Monday revealed the programme was now expected to cost the government five times the original forecast, in real terms, up from £226mn in 2016 to £1.3bn today.

They added:

Meanwhile, the report criticised HMRC for presenting an “incomplete and inaccurate” picture on the initial upfront costs taxpayers would face getting set up on the new system.

This is not news. I presented evidence on Making Tax Digital to a committee of the House of Lords in February 2017, and submitted a 10,000 word submission on the issue. I had this to say back then under the title How and why did HMRC get the Making Tax Digital numbers so wrong?:

I have already noted the report I have written in preparation for giving evidence on HMRC's Making Tax Digital (MTD) programme to the House of Lords Economics Affairs Committee this afternoon. That report is here: because it is 10,000 words a summary has gone to the Lords. Three things stood out whilst writing it from Thursday to Saturday.

The first was the sheer arrogance of the HMRC impact assessment on MTD. This is indicated in a number of ways. There is, for example, no cost estimate made of HMRC's expenditure on this programme when such expense will obviously be incurred. How did HMRC think this would not be noticed? Then there is the glib claim that business will benefit from MTD because of the value of the data generated. The way in which this benefit will arise and be valued is not explained. It is just said that it has been prepared using a standard cost model: this is simply not good enough. Equally arrogant is the assumption that software might replace accountants for some taxpayers. HMRC should really understand why people use accountants in that case, because it's not to add up the books.

Second though was what I can only describe as the incompetence of the numbers presented. HMRC say 5.9 million business will submit four extra tax returns a year (let's not beat about the bush: that is the reality of this programme). That's 23.6 million submissions. And they say they will cost a maximum of £170 million. Divide one by the other and that is £7.20 each, or less than an hour of time at national minimum wage. But HMRC says thew £170 million covers all extra accounting and software costs. I took a strictly marginal approach to costs here: I ignored all accountancy costs and simply looked at how many additional businesses might need new software and estimated it might be 925,000 at £72 a year, the cost of a Quickbooks subscription,  or £67 million a year. That reduced the amount available to cover the cost of submission to £4.36 a return, or 35 minutes of time to undertake these tasks that are bound the necessary as we can be sure that  penalties will be imposed for false declarations:

1. Check cut off for all transactions i.e. make sure a strict allocation to quarters has taken place;

2. Reconcile bank accounts to ensure completeness and correct period end cut off;

3. Check sales ledger completeness and closing debtors to ensure income is appropriately stated and make necessary adjustments e.g. for bad debts if using an accruals accounting method;

4. Check the purchase ledger and outstanding creditors and make necessary adjustments;

5. Accrue period end expenses;

6. Calculate period end prepayments;

7. Count or estimate value of stock and work in progress;

8. Estimate depreciation;

9. Make adjustments for:

   1. Private use of car and other assets;
   2. Use of house as office;
   3. Stock taken as drawings.

10. Ensure that the resulting accounts make sense and correct for errors;

11. Check that drawings are appropriately recorded in a self-employment;

12. Check digital entries prior to submission to HMRC;

13. Log on and submit to HMRC;

14. Check resulting communications from HMRC;

15. Act in accordance with HMRC instructions e.g. re adjusted payments owing.

Even in the case of a simple business these tasks may well take half a day to complete. They may take much longer in a larger business. That is why I call the HMRC impact assessment incompetent.

Third, I wondered how anyone failed to notice that this estimate was so bad before the report was issued? Or did they notice and it went out anyway? Was political expediency the order of the day, and the instruction was to simple have a big enough number in the hope no one would question it? £170 million is enough to daunt most people: their calculators haven't got enough noughts on them to handle it as a friend joked to me last night, and that means most people can't comprehend it. If so, the hope was clearly inappropriate.

But it's also reckless. In addition to an average cost of £305 per small business, as I note in the report, MTD:

1. Potentially imposes inappropriate and potentially seriously misleading accounting methods on taxpayers that may be seriously harmful to their business interests to suit the requirements of HMRC. Business failures as a result of the use of inappropriate accounting data are likely to increase as a result;
2. Potentially threatens the cash flows and so viability of some small businesses in the UK;
3. Threatens the credibility of small business accounting in the UK;
4. Potentially harms UK economic performance as a result.

It is hard to think of a policy more antithetical to the culture of enterprise in the UK than MTD as HMRC propose to impose it. Doesn't this matter to them? And if not, why not?

I was angry in 2017. It was obvious that HMRC were not only way out of their depth on Making Tax Digital, they were also not telling the truth. In my book that, from a tax authority, is unforgivable.

It is now clear that I was right. Making Tax Digital is the failure I predicted it would be, at the much higher cost than was forecast by HMRC, as I predicted.

So, what to do about it?

First, stop the roll out.

Second, cancel any idea of ever extending the scheme.

Third, sack those who lied.

Fourth, rethink the whole issue of what the accounts of small businesses should disclose, including for tax, for which they are hopeless at present.

Fifth, redesign the whole tax return on the basis of maximising the chance of identifying risk.

None of this would be hard, but it would save a fortune and increase tax yield.

Why wouldn't HMRC want to do that?