Wendy Bradley seeks out the truth behind the figures quoted by HMRC for the costs and savings from making tax digital (MTD) for small businesses.
From my conversations with self-employed friends (I am not an accountant), I conclude they generally record takings carefully, because they know they could get into trouble if they fail to record any money coming in. But the business owner may be a bit less careful about recording every penny going out. At tax return time, instead of searching for every last receipt, they include an estimate for say, printer paper and hit "send".
If you have enough money in hand to pay the tax figure the computer spits out, then who cares if you could have paid a bit less tax so long as the thing is done?
What if there are other small traders who are careful with income but less so with expenditure? Will the government really get an extra £625m in tax after 2020-21 from introducing the MTD digital record keeping requirements? Are small business accounts really inaccurate by hundreds of millions a year?
The MTD impact assessment suggests there will be a £400m benefit to businesses in administrative burden savings, but how does this interact with the tax figure?
If there are savings in time spent by the business owner by using digital record keeping, that time can be ascribed a cost which results in a notional saving. A balanced assessment of costs and savings should also consider hard costs like the purchase or license of software and increased expenditure on book-keeping.
These costs, as well as contributing to the admin burden methodology, also contribute in accounting terms to the reduction of taxable profits. If the savings from the £400m MTD admin burden aren't captured in taxable profits but the costs are, then won't the overall taxable profit figure, and thus the overall tax figure, in fact go down?
Calculating the figures
In the main MTD consultation document the tax figure is found on page 67 under “Exchequer Impact”. It says there will be an extra revenue of £10m in 2018-19, £310m in 2019-20 and £625m in 2020-21 and "each year thereafter".
The impact assessment chapter of the same document says the figures have been certified by the OBR and are also published in the Budget (table 2.2 of Budget 2016). Downloading table 2.2 simply gives you an excel spreadsheet which includes the same figures as the draft TIIN chapter.
Looking further at the rubric under the exchequer impact figures in the TIIN, however, we see this statement: "Revenue lost to HMRC due to errors and failure to take reasonable care was estimated at £6.5bn in 2013/14. The methodology behind this estimate was published in Measuring Tax Gaps 2015."
Source of tax gap figures
Measuring Tax Gaps 2015 is archived but still available and it is an interesting read. Figure A.2 in the introduction shows which of the five methods of measuring the tax gap applies to which tax and taxpayer grouping.
The five methods are data matching, top down methods, management information, random enquiries, and illustrative (they are explained further in the document). The MTD population of "SA business taxpayers" is coloured brown to indicate the tax gap has been calculated using "random enquiries".
What are random enquires?
As part of their enquiry programme, as well as the usual "risk based" tax enquiries, HMRC carries out a certain number of "random" enquiries each year.
The exact number is not disclosed, and the taxpayer under enquiry is not told that this is the reason for the enquiry, although all taxpayers subject to enquiries are supposed to be made aware that this is a possibility. It is generally felt that the number of random enquiries is small, perhaps 100-200 per year, although HMRC has assurance from the National Statistics Office that their methodology is robust.
However, at paragraph B15 from the methodological annex it says: “For direct tax estimates based on random enquiries, an adjustment is made to account for under-declarations that are not detected. HMRC continues to undertake analyses to define suitable ranges for other systematic biases in the direct tax estimates. “
In other words, after they have looked at a random enquiry, HMRC add on an "adjustment" to account for anything else they might not have discovered. (The number they first thought of, perhaps?)
MTD will be one of the biggest changes the tax system has ever seen, and it is therefore important that the reasons for making the change are robust and evidence based. I have contacted HMRC on behalf of AccountingWEB, and made a request under the Freedom of Information Act for details of the computation, and the underlying data resulting in the exchequer impact figures in the MTD impact assessment.
Watch this space: The Freedom of Information Act gives HMRC just 20 working days to respond.