The law firm says unannounced visits to premises fell from 669 in 2016/17 to 471 last year, as HMRC’s investment in technology reduces the need for time-consuming and costly raids. There were 761 such raids in 2015/16.

HMRC recently announced that it is experimenting with AI and machine learning for use in ‘complex tax cases’ and compliance; for example, using AI to analyse the likely accuracy of tax returns. HMRC also plans to use AI to automate ten million routine processes by the end of 2018, including customer service and administration functions.
The tax department’s Connect database collects data from multiple public and private sources; including banks, local councils and even social media, to build a complete picture of taxpayers’ spending habits. Sophisticated algorithms then cross-reference information on tax returns with data collected by Connect to flag up names of individuals and businesses for investigation.

HMRC’s ability to identify tax evasion cases has been further strengthened since it started receiving data through the Common Reporting Standard from foreign jurisdictions about UK residents with offshore bank accounts. HMRC has been receiving data from the Channel Islands, BVI and Bermuda since September 2016 and will start receiving data from Switzerland and the UAE later this year. 

Steven Porter, partner at Pinsent Masons, said: ‘HMRC’s big brother-style data collection on taxpayers is giving it the material it needs to ramp up its tax investigations and at the same time, is reducing the need for it to actually raid properties.
‘Falling numbers of property raids by no means indicates a drop-off in activity. HMRC’s growing emphasis on data collection means it now has increasingly large pool of information from which to draw on when launching future investigations.’

Reported by Pat Sweet