Welfare policy in the UK today: the government's flawed approach to fraud

After a consultation held in 2021 to which Privacy International responded, the government has now decided not to expand the powers of its "National Fraud Initiative". However, recent changes in the UK welfare landscape point to troubling plans for the future.

Key findings
  • The UK government launched a consultation around “widening the National Fraud Initiatives' data matching powers.” Privacy International was one of four public interest groups which responded to this consultation. In August 2022, the Government announced that it would not extend the Cabint Office's data matching powers.
  • While this was welcome, recent developments show that the UK government's approach to welfare policy signal an intention to continue approaching welfare policy primarily through narratives of 'cracking down on fraud'.
  • The launch of the UK's new "Public Sector Fraud Authority" is an example of this.
  • Subjecting communities in vulnerable situations to ever-evolving, poorly scrutinised and little publicised digital tools to facilitate their surveillance amidst a cost of living crisis is far from a vision of the economy that aims to support those most in need.
Long Read

In the UK, successive government ministers and members of parliament have made emotive proclamations about the malaise of "public sector fraud".
This year, former Work and Pensions Secretary Therese Coffey said that the welfare system "is not a cash machine for callous criminals and it’s vital that the government ensures money is well spent...[and] fraud is an ever-present threat."
In 2013, the UK's minister for the disabled made numerous claims that there were "vast numbers of bogus disabled [people] who carry on claiming disability assistance long after they have ‘healed’."
Back in 2011, then-Prime Minister David Cameron made a speech to introduce his government's Welfare Reform Bill, including proclamations such as, "[t]he benefit system...allow[s] people to act irresponsibly, [and] often actively encourages them to do so," and "there are totally ineffective sanctions for those who are out to take what they can get and no sense of proportion on what it’s reasonable for people to receive."
Despite repeated references to the risk of welfare fraud and the need to curb it, information as to the specifications and functioning of the technology used to monitor fraud cases remains scarce. Against this background, the launch of a consultation by the Cabinet Office in 2021 surrounding the National Fraud Initiative and the publication of related information, provided an unexpected window into one of the several mechanisms relied upon by the government to detect instances of fraud. You can read more about our understanding of the National Fraud Initiative here.

In this article, we analyse the outcome of the National Fraud Consultation against a backdrop of recent reforms pursued by the UK government which indicate a shift in fraud detection/prevention powers and functions in the public sector. We argue that conflating all types of fraud against the public sector - including tax fraud, corporate fraud, fraud in public procurement and fraud committed by organised criminal networks - with what is commonly known as "welfare fraud" negatively impacts the right to social protection in the UK and raises serious concerns around economic justice, equality and privacy.

The 2021 National Fraud Consultation: a positive result

In 2021, the UK government launched a public consultation on extending its data-matching powers through new legislation.
Data-matching broadly refers to a power (which is currently exercised by the Cabinet Office) to "compare sets of data to determine how far they match." Currently, the National Fraud Initiative (NFI), which is 'housed' within the Cabinet Office, obtains data from public sector bodies that are required to submit data to it on a regular basis (including, for example, local authorities and pension schemes), as well as private sector bodies which submit data voluntarily. Public and private sector bodies can then make use of the NFI's "data-matching" service in order to prevent or detect fraud. The current legislation makes it clear that this power can only be exercised to assist in the prevention and detection of fraud.[1]
Privacy International responded to the consultation that was put out in 2021, arguing that expanding the purposes for which the "data-matching" power could be exercised posed a threat to privacy, the fundamental rights of data subjects and raised issues related to fairness and discrimination. Importantly, we also highlighted that expanding data-matching powers to include debt-recovery would amount to disproportionate sanctions for people facing poverty.
In August 2022, the Government decided not to proceed with extending the NFI's powers to new purposes "at this stage". In it's response, it acknowledged that "[m]any respondents [to the consultation] were critical, and the majority of those that were critical emphasised the need for appropriate safeguards and mitigations to protect individuals and their privacy." The government also stated that it "[understood] the concerns expressed by participants relating to data access, use of data, and privacy."
The consultation outcome, while encouraging, did not spell the end of the National Fraud Initiative. Notably, the government later announced that a "new Public Sector Fraud Authority" would house the work of the National Fraud Initiative, and that this new authority would continue to support public bodies in "enhancing their fraud response through the use of data and analytics."

2022: the Cabinet Office bows out, enter the Public Sector Fraud Authority

This year, another player joined the fight against public sector fraud: the brand new "Public Sector Fraud Authority" (PSFA), which is ostensibly intended to replace the group in the Cabinet Office which housed the National Fraud Initiative.
Officially announced in August 2022, the PSFA's mandate document describes the public body as the "UK government’s Centre of Expertise for the management of fraud (and associated error) against the public sector", and as the authority which "leads the Government's Counter Fraud Function".[2] Some of the services the PFSA is intended to deliver include "data intelligence services...by providing data analytics services" to ministerial departments and public bodies, and "running a unit to coordinate, facilitate and share intelligence across central government."
Concerningly, a PSFA brochure indicates that it will introduce a National Counter Fraud Data Analytics Service (NCFDAS) which will provide "best in class tools and techniques" to find and prevent fraud, "build[ing] on the long standing National Fraud Initiative". At the time of writing, there is no public information available about the NCFDAS.

Extract from PSFA brochure.

While the PSFA is intended to provide support services to government institutions, it will not substitute independent departmental efforts to curb fraud.
Around the same time that the PSFA was announced, the UK's Department for Work and Pensions (DWP) published their plan to "fight fraud in the welfare system", arguing that a plan to invest £613 million in "frontline counter-fraud professionals and in enhanced data analytics" would stop billions of losses resulting from fraud and error over the next three years. This plan included passing legislation that would give civil servants increased powers to access to data from third parties "and carry out arrests and searches and seize evidence."
The legal landscape in the UK is also evolving in a direction likely to facilitate data processing of the nature indicated by the PSFA. Draft changes to data protection legislation in the form of the Data Protection and Digital Information Bill are already seeking to water down GDPR safeguards against automated decision-making.
It is against this back-drop that we saw the UK Government retreat from its plans to widen the existing "data matching powers" of the UK's National Fraud Initiative. Contrary to the expectations set by the consultation outcome when analysed in isolation, the UK government powers and solutions to counter fraud are far from static, and continue to be expanded.

Getting the basics right: fraud against the public sector vs. "benefits fraud"

Conversations about fraud tend to conflate benefits fraud with fraud against the public sector at large.
Public sector fraud is a broad term which encompasses a range of issues with public expenditure. It encompasses fraud, bribery and corruption which targets or leads to losses of public funds, as well as 'error' associated with the expenditure of public funds. On the other hand, "benefits fraud" relates to instances where individuals fraudulently represent that they are entitled to claim certain social protection payments from the UK's Department for Work and Pensions.
While there is evidence which suggests that the Government should pursue policies to tackle public sector fraud in general, the focus remains on benefits fraud - and recent developments suggest that the approach taken by government is likely to be the same regardless of the type of fraud at stake. Crucially, nothing in the PSFA announcement or brochure differentiates between types of fraud that must be tackled.
There is a simple, yet powerful reason for benefits fraud to be addressed separately from corporate and tax fraud, or fraud committed by organised criminal networks. Social protection payments exist to support vulnerable members of our community. There is a human rights imperative behind social protection payments: not only is the UK compelled to uphold peoples' fundamental right to live with dignity, but more concretely it must ensure that it protects and enforces the right to social security as outlined in the International Covenant for Economic, Social and Cultural Rights, to which the UK is a signatory. These obligations should be the starting point to assess the necessity and proportionality of any initiative that is aimed at tackling ‘benefits fraud’. However, as early as 2018, experts were sounding the alarm about the UK's approach to benefits. As of 2022, these concerns continue to be expressed.
Policy initiatives that seek to tackle fraud while protecting the right to social security and dignity will - and should - be significantly different to, and quite apart from, initiatives to tackle organised crime, corporate fraud, and tax fraud.
Using the same tools is not necessary, nor is it proportionate. The impact of surveilling and criminalising people who are accessing life-saving support through excessive reliance on data analytics can have a devastating impact on people's lives, and raises serious concerns around economic justice, equality and privacy. People should not be penalised through surveillance and quasi-criminal sanctions for struggling with unemployment or poverty.

Challenging the narrative around welfare fraud in the UK

Welfare fraud may be a pressing issue, but it is by no means the sole or most prominent source of risk for the government - and according to experts, current approaches to combatting fraud would appear to be short-sighted.
There is more to fraud than welfare fraud
A parliamentary report prepared by the cross-party Committee of Public Accounts in 2021 addressing the need for government to tackle "fraud and error" found that the government could do more to ensure that it was tackling fraud holistically. The report urged the government to improve its methods of tackling fraud specifically outside the benefits and welfare system, stating that "for many areas of spend outside the tax and benefits system there is still no formal measurement and limited capability to tackle fraud and error," and that "[g]overnment’s focus on the long-standing fraud and error risks it understands, risks large amounts of fraud and error being unidentified or untackled elsewhere." Additionally, the report made it clear that government departments are lacking "robust measurements of fraud and error". These shortcomings must be addressed prior to implementing additional controls or expanding surveillance and data-collection powers.
Terminology matters
As stated above, all too often public sector fraud is conflated with benefits fraud - including by government.
For example, the UK Government's "plan to fight fraud in the welfare system" introduces "the nature and scale of fraud" by stating that, prior to the pandemic, "the cost of public sector fraud was £29.0 billion." It is noteworthy that a document setting out a plan to fight welfare fraud, which is only a subcategory of public sector fraud, would cite figures pertaining to public sector fraud at large.
What the document does not mention is that this number includes government error and all types of fraud including, for example tax fraud and fraud against the ministry of defense. Additionally, the DWP's final estimate which states that there was £8.4bn of over-payments to claimants by the DWP includes over-payments by reason of government error.
In reality, there is a significant difference between the cost of welfare fraud and the cost of public sector fraud. A 2021 government bulletin noted that the estimated fraud cost to the government outside of the tax and welfare system was between £2.5 and £25bn per year. This estimate increases to £29.3bn-£52bn when fraud against the tax and welfare system are included. Applying those estimates to the most recent government figures of welfare fraud (FY 2021 and FY 2022), even when using the lower end of the government's overall fraud range, the value of welfare fraud would amount to less than one fourth of the overall fraud cost. Using the higher end of the government's overall fraud range, the value of welfare fraud would amount to one eighth of the total fraud value.
How a fraudulent claim is identified
There are valid questions as to whether one can take reported instances of welfare fraud at face value.
The DWP will deem a claim to be fraudulent when claimants "fail to provide evidence" or "fail to engage fully" in the review process, and then give up their benefit entitlement. Nevertheless, policy-makers argue that "updating the [DWP's] powers to boost access to data from third parties and carry out arrests and searches and seize evidence" and pushing for "more flexible and proactive use of data," is necessary because of the huge cost of public sector fraud to the public.
Additionally, the Public Sector Fraud Authority’s mandate document estimates that public sector fraud leads to a loss of £33bn a year – an estimate which once again, includes statistics related to government error and a range of fraudulent activity. Nevertheless, when the PSFA was launched, a spokesperson from the treasury department said that the new authority would "renew our efforts to combat people taking advantage of our public services and support...[and] reinforce wider investment in government to crack down on fraud and mis-claiming.”
While the National Fraud Initiative consultation outcome is overall positive, recent changes show that the UK government continues to develop its fraud prevention and detection capabilities - details of which remain forthcoming.
We do not think that the government should continue to expand its powers to monitor people who access public services by collecting and analysing their personal data, sharing it with third parties, and contracting with companies like Palantir to provide 'data anayltics' under the guise of fighting public sector fraud.
Subjecting communities in vulnerable situations to ever-evolving, poorly scrutinised and little publicised digital tools to facilitate their surveillance amidst a cost of living crisis is far from a vision of the economy that aims to support those most in need.
We will continue to investigate and challenge policies which expand public bodies' - such as the PSFA's - powers to collect and analyse personal data in ways which subject our communities to surveillance and sanctions simply for accessing public services.


[1] This is a statutory power under section 33 and schedule 9 of the [Local Audit and Accountability Act 2014](https://www.legislation.gov.uk/ukpga/2014/2/schedule/9/enacted#schedule-9-paragraph-1-1).

[2] We understand that the Government's [Counter Fraud Function](https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1014385/6.7628_CO_Govt-Functional-Std_GovS013-Counter-Fraud_v4.pdf) was previously part of a broader effort by the UK Government to implement what it has called ["Functional Standards"](https://www.gov.uk/government/collections/functional-standards) for doing business across government bodies. According to the [most recent update of the Funcational Standards (June 2022)](https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1092024/GovS-001-government-functions-WEB-version.pdf), the Counter Fraud Function is no longer listed as a stand-alone Functional Standard. Instead, that the government has is [pressing ahead with plans](https://www.icaew.com/insights/viewpoints-on-the-news/2022/apr-2022/funding-concerns-over-new-public-sector-fraud-authority) to turn its counter-fraud function into The Public Sector Fraud Authority.