HMRC's newly published research report, Evaluating additional tax revenue from Making Tax Digital for VAT, creates a mathematical model for proving the success rate of MTD for VAT in its first years. It concludes "There is a high likelihood Making Tax Digital is generating additional tax revenue..." The methodology behind the research is pure hogwash and you don't need maths to reach this conclusion.
Comment by Nichola Ross Martin CTA (fellow) FCA
One of the aims of Making Tax Digital (MTD) for VAT is to reduce errors. At the outset, HMRC gave us various estimates of the tax cost of those errors and those estimates were as murky as the precise total of 'the tax gap'. The tax gap is the difference between the theoretical amount of tax that should be paid and the actual tax receipts. Tax errors are a bit of a 'Rumsfeld', i.e. known unknowns, we know that they happen but we cannot accurately quantify them.
HMRC's new research into the success of MTD for VAT in correcting errors attempts to accurately quantify the benefits of MTD. It's about as useful as a chocolate teapot, much more expensive to create and does not really prove anything more than we already know.
One might have thought that the best way to try and quantify tax errors would be to set a control trial of a sample of businesses and then set them tests and see how they perform measuring their error rate before MTD and then after using MTD software. It would not be easy, as of course as we all know, all businesses are different and not all errors are repeated but it would at least give some statistics on human error as a starting point.
A control trial was not used in this research and instead, the researchers created a complicated statistical model: "Propensity score matching in conjunction with a difference-in-differences model were used to estimate the additional tax revenue from the introduction of Making Tax Digital for VAT taxpayers who file their returns quarterly". This part relies on input variables which are in turn given a weighted score as a kind of risk rating and that was then heavily tweaked. Variables stemmed from "Various business characteristics and behaviours, such as whether the business already uses software and which industry they trade in". Once a model was created the researchers ran data through it, made more adjustments, although we are unsure what and then you get an output that is presumably in some kind of expected range. "The difference-in-differences estimator is the isolated average change in tax declared due to Making Tax Digital, controlling for inherent biases between groups as much as we can."
This led to Equation 1 – Difference-in-differences equation
DiD = (T1 - T0) - (C1 - C0)
I won't go on, I can see your eyes are glazing up. To cut to the point, we learn from that model that MTD is producing additional revenue per business of £57 per quarter.
The researchers conclude,
“Overall, we conclude that there is a high likelihood Making Tax Digital is generating additional tax revenue by reducing errors and making it easier for businesses to get their tax right.”
I do not want to be rude about this kind of research, but I could put a finger up to the wind and come out with a statement like that. Surely almost anyone in business can work out that if you remove a human from certain elements of bookkeeping you are bound to be better off: machines add up correctly and they don't make transposition errors.
What did the research miss?
The research was unable to cost the results as to whether people were charging or claiming VAT correctly though. For all we know, many of the VAT-registered traders using software for the first time will accept what their software suggests in terms of VAT rates, exemptions etc regardless of the VAT law. We can only hope that traders who are not assisted by accountants were able to set their software codes to cash accounting rather than accruals accounting and that those who have them have actually programmed their tills/iPads correctly in the first place.
I am eagerly anticipating the outcome of the research on MTD for the lock-down years. How will the researchers tweak their algorithms for the fact that many businesses have stopped taking cash, which should boost tax revenue (if your bank feed is set up correctly)? Will the effects of the change in the VAT rate for the hospitality sector be an added weighted factor in the model? It seems that many businesses were having to estimate turnover for the reduced rate calculation.
What about the cost of computer errors?
As an end note, the researchers have also not factored in a new issue. I know, from bitter experience, that if you make an error when using software, it can be incredibly hard to correct it again. As one of HMRC's officers will attest, it took us something like seven VAT quarters to iron out issues caused by recording keeping through the change in the hospitality rate and some five different sets of calculations on a spreadsheet to reconcile the VAT account. It was like the blind leading the blind as HMRC's Agent Sevices Account does not record or reconcile repayments and overpayments and neither side knew what the other could or couldn't see.
Read it through: HMRC's research Evaluating additional tax revenue from Making Tax Digital for VAT
The views in this article are just that, views made by the author Nichola Ross Martin CTA (fellow) FCA
Join thousands of accountants and advisers (and their clients) who use www.rossmartin.co.uk as their primary TAX resource.
Register with us now (for free😍) to receive our FREE weekly SME Tax News update, tax tips and tools.