Rogue Landlords

5th August 2020

Landlord tax evasion costing up to £1.7bn a year


Tax evasion by residential landlords could be costing the Treasury up to £1.73 billion a year, a new analysis by TaxWatch has found. This is more than three times the amount reported for 2010, the only time official data has been published on the scale of tax evasion in the whole of the buy to let sector. This is despite HMRC running a high profile campaign to deal with the issue over a number of years.

The size of the problem, and the failure of the existing system to get to grips with it suggests a new approach is required. We recommend that HMRC make better use of the data available to them to identify residential landlords, and consider setting up a national database of rental properties, issuing each rental property with a unique number so that HMRC is able to see which properties are being rented out and by whom.

Estimating the buy-to-let tax gap

In 2013 the Government estimated that up to 1.5 million landlords had underpaid or failed to pay up to half a billion pounds in tax for the 2009-2010 financial year.12

This appears to be the last time that the government completed an estimate of the scale of tax evasion in the buy-to-let sector. On 13 July 2020 Stephen Timms, the Labour MP for East Ham, asked the Chancellor of the Exchequer what the most recent estimate is for tax lost through landlords failing to declare their income.

The Financial Secretary to the Treasury responded, stating:

“The information requested is not available as HM Revenue and Customs (HMRC) does not make a separate estimate of the proportion of the total tax gap attributable to residential landlords.

HMRC does estimate the tax gap arising from individuals in employment who have not declared and therefore not paid tax on lettings income. The latest estimate of this tax gap was £540 million for the tax year 2018-19.”3

We contacted HMRC to ask for clarification as to what ‘in employment’ meant exactly, and were told that it means those who pay tax on their earning from their main job through Pay As You Earn (PAYE).4

However, the vast majority of tenancies are held by people not in employment according to the HMRC definition. We also know that tax evasion amongst self-employed landlords is rife. According to research conducted by Arun Advani of the University of Warwick, one quarter of landlords that file self assessment returns do not declare all their income, on average not reporting 60% of their property income.5

The English Private Landlord Survey 2018 shows that at least two thirds of tenancies in England are owned by those who will not be receiving pay through PAYE.6 What this means is that the £540m in unpaid tax is attributable to approximately only a third of the tenancies at most. If the tax behaviour of the two thirds not ‘in employment’ is similar to those who are, and we have no reason to believe otherwise, the total tax gap is then likely to be as much as £1.73bn.

HMRC’s previous attempts to deal with the issue

In 2013 HMRC began their ‘Let Property Campaign’. The campaign was initially set to run for 18 months although it has now been extended indefinitely. The idea was that it would allow landlords to ‘get up to date with their tax affairs in a simple way and take advantage of the best possible terms.’7

The campaign sees threatening letters sent to landlords encouraging them to voluntarily disclose that they had not paid the full amount of tax on their income. If there is no response within 30 days of receiving these letters, landlords are liable to face penalties for non-compliance.

The approach seems to be having little impact. Data obtained by accountancy firm Saffery Champness through a Freedom of Information (FOI) request in 2019 showed that in the five years since the campaign started, just 35,099 people had made voluntary disclosures to HMRC, only 2.3 per cent of the individuals originally identified. This number falls far short of the estimated 1.5 million landlords who were underpaying or failing to pay tax in 2009-2010.8

Big Data

One part of the problem is finding out who is a residential landlord.

One possible solution could lie in a UK landlord database, an idea which has been advocated for some time. Scotland, Wales and Northern Ireland already have landlord registration schemes, but England does not. This would allow HMRC to see who is a landlord and quickly check if the landlord is declaring property income on their tax returns, or if they are filing a tax return at all.

To see how this could work in practice, the borough of Newham in east London serves as a good example. Newham Council introduced a compulsory borough-wide licensing scheme for landlords in 2013. Before the scheme started the borough estimated that Newham had 30,000 rental properties let out by 5,000 individual landlords. It discovered that there were in fact 50,000 properties let by 27,000 individual landlords. Furthermore it found that 13,000 landlords, almost 50% of the total, had not registered for self-assessment on their rental income. Newham provided the data on these landlords to the government and HMRC has said that their work with the borough has resulted in the recovery of £115m in unpaid tax.9

The benefits of the scheme go beyond an increase in income tax receipts. Newham Council have said that their licensing scheme has allowed them to increase council tax receipts by £3.1m.10

The government has been extremely reluctant to roll out licensing schemes used by boroughs like Newham nationwide, deeming it to be an unnecessary regulatory burden on landlords.11 When asked on 13 July if the Government had made any estimate into what additional tax receipts would be raised from a national licensing scheme for landlords, the answer given was simply that landlords are liable to income tax.12 The minister for Housing, Communities and Local Government confirmed that “The Government has no current plans to introduce a national licensing scheme and as such it is unable to offer any estimate related to such a scheme.”13

However, even if the government has no intention of introducing a national landlord licensing scheme, there is much more that could be done.

Firstly, the government could use the data that already exists more efficiently. There are plenty of data already available – the land registry, reporting from estate agencies, tenancy deposit schemes – but these are not stored in a single record, rather, they are only viewed during individual HMRC investigations, a method which clearly isn’t working. In the 12 months to June 2019, 234,000 buy-to-let mortgages were taken out.14 There could be a requirement on lenders to provide data to HMRC on who owns a buy-to-let mortgage.

One further step could be to create a central record of rental properties (not landlords), with each property given a unique reference akin to a National Insurance Number. This could be cross referenced with the land registry to identify the property owners and assist HMRC in collecting tax from landlords. Having this number could be a requirement of letting out a property and obtaining a buy-to-let mortgage on a property. It would be a significantly lower regulatory burden than a full licensing scheme for landlords.

The solution to the significant issue of tax evasion in the rental sector is out there, should there be a will to tackle it.

2Pay your tax or face the consequences, minister tells buy-to-let investors, The Telegraph, 19 September 2013,

4Correspondence between HMRC and TaxWatch, 29 July 2020.

5Who does and doesn’t pay taxes?, Institute for Fiscal Studies,

6The English Landlord Survey allows people to categorise themselves more than once, which means that the categories of landlords add up to more than 100% of the total tenancies. In order to come to our figure of landlords not “employed”, we have removed anyone who classed themselves as ‘full-time employed’ or ‘part-time employed’. The total tenancies owned by these two groups amounts to 31.2 per cent. If you subtract 31.2 per cent from a total of 100 per cent, you are left with 68.8 per cent of tenancies not belonging to those “in employment”. In reality, it is likely that some people in the part time employed category in the HCLG survey will have a main employment which is not paid under PAYE, and so they will not be included in the HMRC data. This means our estimate of how many landlords and tenancies are in employment under the HMRC definition is likely to be larger than the real amount, which would make our figures an underestimate of the problem.

8HMRC’s targeting of accidental landlords falls short, FT Adviser, 14 March 2019,

9London landlords ‘avoiding £200m in tax’ by not declaring income, The Independent, 14 August 2017,

10Newham finally wins right to carry on with licensing scheme – after a fight, Property Industry Eye, 05 December 2017,

11On 05 November 2019, in response to a question on a national landlord register, Conservative peer Viscount Younger of Leckie stated that a register could “place an additional regulatory burden on landlords”,

14Buy-to-let statistics, Finder, 07 February 2020,