Fraud and Error in Welfare Benefits, 2016 to 2022: British Social Attitudes survey

Question 1

Voluntary statement of compliance with the Code of Practice for Statistics

Accepted Answer

The Code of Practice for Statistics (the Code) is built around 3 main concepts, or pillars, trustworthiness, quality and value:

trustworthiness – is about having confidence in the people and organisations that publish statistics
quality – is about using data and methods that produce assured statistics
value – is about publishing statistics that support society’s needs for information

The following explains how we have applied the pillars of the Code in a proportionate way.

Trustworthiness

The analysis has been carried out internally. Data tables have been included for transparency. Analysis complies with expected standards of the GSR Code. Although research findings are shared with ministers and other officials before publication, this is done to promote the value of the research to the department. Ministers have no editorial role.

Quality

Analysis was carried out using established quantitative research methodology and statistical methods. Analysis was conducted for each individual year of data. Any differences reported between years of data in the time series have been tested for statistical significance. Each stage of analysis and reporting has been quality assured using internal quality checking processes.

Value

Findings from the research will supplement our wider evidence base on public perceptions of fraud, error and debt.

Question 2

Executive summary

Accepted Answer

Overview

The British Social Attitudes (BSA) survey is an annual survey, conducted by the National Centre for Social Research. In 2022, 6,638 fully productive surveys were completed, with questions funded by the Department for Work and Pensions (DWP) asked to a subset of respondents (versions 7 to 12), giving a sample of 3,310 people in private households, aged 18 and over. Addresses were selected at random, inviting up to two household members to take part, to provide a representative picture of the attitudes of the British population.

DWP is one of a number of funders of the survey and each year includes different modules of questions. This report covers attitudes to benefit fraud and error, using questions asked in 2016, 2017, 2018, 2019 and 2022. Due to the pandemic, there was a gap in the survey, with no questions asked on fraud in 2020 and 2021. Some questions on fraud have been asked every year, whereas others have been asked less frequently. Repeating some questions over time provides a unique insight into how social attitudes have changed over time.

The survey questions aim to understand attitudes to benefit fraud and error in Britain, as well as perceptions of the government’s efforts to tackle benefit fraud. The purpose of the research was to inform DWP’s evidence base on public attitudes to benefit fraud and error.

Research Context

BSA is an important evidence source for monitoring public attitudes to benefit fraud and error over time, which we have not previously published. The research aimed to explore the British public’s perceptions of the government’s efforts to tackle benefit fraud, to help inform future policy development, assess communication approaches and build the evidence base on fraud and error.

The research also sought to understand public perceptions of fraud and error in the benefits system and how this may have changed over time. The survey has been conducted to the backdrop of various changes within society including: the UK leaving the European Union; the COVID-19 pandemic; and political change. The number of people claiming Universal Credit stood at 5.6 million in January 2022, a substantial increase since 2016.

Methodology

The British Social Attitudes survey is designed to yield a nationally representative sample of adults aged 18 or over.

The report focused on questions asked in 2016, 2017, 2018, 2019 and 2022. The report chose to begin the time series analysis from 2016 as it provided the best comparison with later years of data. Percentage change between years of data have been tested for statistical significance and where a difference has been highlighted, this difference is significant at the 95 percent level. Analysis has largely compared responses from 2016 with responses from 2022 to indicate changes in social attitudes from the earliest point in the time series to the most recent data collection.

BSA was administered as a face-to-face survey until 2020, when due to the pandemic, the survey was adapted to push-to-web, with no face-to-face interviews as in previous years. Differences between modes of data are addressed with weighting although some bias inevitably remains. Caution should therefore be taken when comparing differences between years of data as differences between years may be as a result in the change to the survey mode of data collection.

Key Findings

In general, the findings suggest most respondents thought benefit fraud was wrong. While this remains a majority response, this attitude has decreased over time and there is a growing tolerance towards benefit fraud.

BSA explored the acceptability of benefit fraud across four different scenarios, the findings are as follows:

Scenario 1: An unemployed person on benefit takes a casual job and is paid in cash. He does not report it to the benefit office and is £500 in pocket. Do you feel that this is wrong or not wrong?

In 2022, 53% of respondents believed it was wrong (“wrong” or “seriously wrong”) if an unemployed person on benefit took a casual job and was £500 in pocket. However, this attitude decreased since 2016 from 68% and instead there was an increase in respondents who reported that this scenario was not wrong (“not wrong” or “a bit wrong”) from 27% in 2016, to 38% in 2022.

Scenario 2: An unemployed person on benefit takes a casual job and is paid in cash. He does not report it to the benefit office and is £3,000 in pocket. Do you feel this is wrong or not wrong?

In 2022, 66% of respondents believed it was wrong (“wrong” or “seriously wrong”) if an unemployed person on benefit took a casual job and was £3,000 in pocket. This attitude has decreased since 2016 from 80%, with an increase in respondents suggesting the situation is not wrong (“not wrong” or “a bit wrong”) from 16% in 2016, to 27% in 2022.

Scenario 3: Suppose someone knowingly gave false information to support their benefit claim. What would your view about this be?

In 2019, 87% of respondents agreed that benefit fraud was wrong (“always” or “usually wrong”). There was evidence of a small decrease from 2016 in which 91% of respondents thought benefit fraud was wrong. There was also an increase in respondents who answered that it was “sometimes wrong”, from 7% in 2016 to 11% in 2019.

Scenario 4: A person on benefit moved in with his girlfriend and her young daughter a year ago he does not report this, and the couple are £3,000 in pocket, is this wrong? Looking at claimants failing to disclose a partner moving in and receiving a £3,000 overpayment, 76% of respondents thought this was wrong (“wrong” or “seriously wrong”) in 2019.

Most respondents thought that the government were not doing enough to tackle fraud and that they were not likely to catch those committing fraud.

BSA asked about efforts by the government to tackle benefit fraud. Overall, respondents thought it was unlikely (“very” or “not at all likely”) that someone who falsely claimed benefits would be caught. In 2022, 61% of respondents had this view, compared with 2016 this attitude had increased (from 54%), however this trend was mixed over that time period.

Similarly, in 2022 60% of respondents did not think the government were doing enough to reduce benefit fraud. There was evidence of a small increase compared with 2017 in which 57% respondents did not think the government were doing enough, however this trend was mixed over that time period. The proportion of respondents with the view that the government were doing ‘too much’ increased from 4% in 2017 to 8% in 2022.

Across four questions the British public had a low awareness of penalties for benefit fraud in 2019, but this had increased overall, since 2017.

BSA also asked about the public’s awareness of the penalties that can be applied for someone committing benefit fraud:

If someone is found guilty of a benefit fraud offence, certain benefits can be reduced or stopped for up to 3 years. Were you aware of this before the interview?

In 2019, 71% of respondents were not aware that if someone was found guilty of benefit fraud, certain benefits will be reduced or stopped, although a sizeable proportion (29%) were aware.

If someone is found guilty of benefit fraud, they are always required to pay back any extra money they were paid. Were you aware of this before the interview?

In 2019, 61% of respondents were aware that someone found guilty of benefit fraud must pay back any extra money they were paid. This was one of the most consistently recognised penalties for benefit fraud, out of the four penalties asked about in the survey. Since 2017, the number of respondents who were aware of this penalty, increased from 54% to 61%.

If someone commits benefit fraud, they can choose to receive a financial penalty of up to £5,000 instead of being prosecuted. Were you aware of this before the interview?

In 2019, 87% of respondents were not aware that if someone committed benefit fraud, they could choose to receive a financial penalty of up to £5,000 instead of being prosecuted. Although, the number of respondents who were not aware of this penalty type has decreased since 2017 (from 92% to 87%), whereas the number of respondents aware has increased (from 8% to 12%).

If someone is found guilty of benefit fraud, they will have a criminal record. Were you aware of this before the interview?

In 2019, 61% of respondents were aware that someone found guilty of benefit fraud would receive a criminal record. This was one of the most consistently recognised penalties for benefit fraud, out of the four penalties asked about in the survey. Since 2017, the number of respondents who were aware of this penalty, increased from 48% to 61%.

The British public had mixed views on the cause of benefit fraud.

Respondents were asked how far they agree with two statements about the cause of benefit fraud:

How much do you agree or disagree that a lot of false benefit claims are a result of confusion rather than dishonesty?

In 2022, respondents had mixed views regarding whether false benefit claims result from confusion rather than dishonesty. Approximately a quarter of respondents agreed (26% “strongly agree” or “agree”) with this statement, whereas roughly a third disagreed or were neutral (32% “disagree” or “disagree strongly” and 34% “neither agree or disagree”).

How much do you agree or disagree that the reason that some people on benefit cheat the system is that they don’t get enough to live on?

Furthermore, in 2022, 39% of respondents agreed that people cheat the benefit system because they don’t get enough to live on whereas 25% were neutral and 30% disagreed. More people agreed than disagreed in each year the question was asked.

Question 3

1.  Background

Accepted Answer

1.1. Context

The British Social Attitudes (BSA) survey is an annual survey that has been running since 1983^{[footnote 1]}. During that time, almost 120,000 members of the public have been surveyed, each year asking up to 6,000 people up to 300 questions about their attitudes on a variety of topics. Respondents are selected at random to provide a representative picture of attitudes of the population.

The Department for Work and Pensions (DWP) is one of a number of funders of the survey and each year includes different modules of questions. This report covers attitudes to benefit fraud and error using questions asked in 2016, 2017, 2018, 2019 and 2022. The survey ran between 2019 and 2022, however there was a gap in fraud and error questions being included as a result of the pandemic. Some questions have been asked every year, others every couple of years, and others less frequently. Repeating some questions over time provides a unique insight into how social attitudes have changed over time. A copy of the fraud and error module from the 2022 questionnaire is included in Annex A at the end of the report. Annex B includes a table which covers the fraud and error questions asked in each year of the time series included in this report.

1.2. Methodology

1.2.1. Sampling

Full methodological details are available in the Technical Reports produced by the National Centre for Social Research who led the British Social Attitudes survey^{[footnote 2]}. The British Social Attitudes survey is designed to yield a nationally representative sample of adults aged 18 or over. Since 1993, the sampling frame for the survey has been the Postcode Address File (PAF), a list of addresses (or postal delivery points) compiled by the Post Office. For practical reasons, the sample is confined to those living in private households. Each year of data has a different sample size and different weighting considerations. The methodological details below are taken from the most recent user guide for 2022.

A sample file of non-clustered addresses was drawn from the PAF. 44,280 total addresses were then allocated to a main sample (N= 36,900) and a reserve sample (n=7,380). The reserve sample is not issued unless it is deemed necessary to meet target interviews. The reserve sample was not issued in 2022. Up to two individuals aged 18 or over at each sampled address were eligible to take part in the survey. The change in methodology from “face-to-face” to “push-to-web” meant that non-response could occur within households as well as at the level of the address. This contrasts with previous years where one adult was selected at random at each address, and selection weights were used to account for differential selection probabilities. Separate non-response models were constructed to deal with each of these elements of non-response. A calibration weighting was also used to adjust the profile of the responding sample to match the population in terms of age, sex, education, tenure, ethnicity, economic activity and region. There was no selection weight or separate self-completion weight due to the change in method, as a result only one weight was available in 2022 (BSA22_final_wt), which was applied to the data before running any analysis.

1.2.2. Base for analysis

The unweighted base is provided in this report for each question, to account for varying sample sizes in each year and the type of questionnaire answered (full survey or self-completion). BSA responses of ‘Don’t know’ or ‘Prefer not to answer’ are considered to be valid responses and are reported in the base for analysis and reported in the analysis tables.

All figures have been rounded to integers, no decimal places are presented.

1.2.3. Changes over time

Prior to the mode change in 2020, BSA required respondents to answer a self-completion questionnaire in addition to the main survey. The rate of self-completion response differed from survey to survey. In 2018, unlike previous years, analysis showed that there were statistically significant differences in the profile of those who returned a valid self-completion survey and those who did not. As a result, it was deemed appropriate to generate a self-completion non-response weight which took account of some of the systemic underlying factors which may have been leading to these differences. This was applied for analysis of 2018 questions.

BSA was administered as a face-to-face survey until 2020, due to the pandemic, when the survey adapted to a mixed-mode design of web and CATI (telephone interviews), with no face-to-face interviews as in previous years. Differences between modes of data are addressed with weighting although some bias inevitably remains. For example, push-to-web surveys tend to overrepresent older people and people with degrees. The design effects tend to be higher for the web survey such as standard errors larger and confidence intervals wider. Caution should therefore be taken when comparing differences between years of data as differences between years may be as a result in the change of survey mode.

Percentage change between years of data have been tested for statistical significance (incorporating the complex sample design of the survey) and where a difference has been highlighted, this difference is significant at the 95 per cent level. Data has largely compared responses from 2016 with responses from 2022 to indicate changes in social attitudes.

Question 4

2.  Attitudes to Fraud and Error in Welfare Benefits over time, 2016-2022

Accepted Answer

2.1. Cause of Benefit Fraud

The British Social Attitudes (BSA) survey asked why people might commit benefit fraud. In particular, respondents were asked:

How much do you agree or disagree with each of these statements?

a lot of false benefit claims are a result of confusion rather than dishonesty.
the reason that some people on benefit cheat the system is that they don’t get enough to live on.

In 2022, respondents had mixed views regarding false benefit claims resulting from confusion rather than dishonesty. 26% of respondents agreed (“strongly agree” or “agree”) with this statement, whereas 32% disagreed (“disagree” or “disagree strongly”) and 34% were neutral (“neither agree or disagree”). Disagreement with this statement has largely decreased since 2016, from 40% to 32% in 2022. Whereas neutrality has increased from 28% in 2016 to 34% in 2022 (Table 2.1).

Table 2.1 – How much do you agree or disagree with each of these statements? A lot of false benefit claims are a result of confusion rather than dishonesty.

	2016	2017	2018	2019	2022
Agree Strongly/ Agree	26%	N/A	28%	26%	26%
Neither agree or disagree	28%	N/A	27%	30%	34%
Disagree/ Disagree strongly	40%	N/A	38%	35%	32%
Don’t know	4%	N/A	4%	6%	7%
Refusal	2%	N/A	2%	3%	1%
Unweighted bases	2,400	N/A	2,269	2,636	3,310

Source: The National Centre for Social Research, British Social Attitudes survey, 2016, 2018, 2019 & 2022

In 2022, 39% of respondents agreed (“agree strongly” or “agree”) that people were cheating the benefit system because they don’t get enough to live on. Conversely, 30% of respondents disagreed (“disagree” or “disagree strongly”) with this statement, and 25% were neutral. There has been an increase in the number of respondents that agreed with this statement since 2016 from 34% to 39% in 2022, and a decrease in those that disagreed from 38% in 2016 to 30% in 2022 (Table 2.2).

Table 2.2 – How much do you agree or disagree with each of these statements? The reason that some people on benefit cheat the system is that they don’t get enough to live on.

	2016	2017	2018	2019	2022
Agree strongly/ Agree	34%	N/A	41%	38%	39%
Neither agree or disagree	23%	N/A	23%	25%	25%
Disagree/ Disagree strongly	38%	N/A	30%	29%	30%
Don’t Know	3%	N/A	4%	5%	6%
Refusal	2%	N/A	3%	3%	2%
Unweighted bases	2,400	N/A	2,269	2,636	3,310

Source: The National Centre For Social Research, British Social Attitudes survey, 2016, 2018, 2019 & 2022

2.2. Prevalence of Benefit Fraud

BSA asked how common benefit fraud is. Since 2016, respondents commonly believed 0-20 people (out of every 100 people receiving benefits) had broken the law by giving false information to support their claim. In 2022, 53% of respondents believed over 20% of claimants had committed benefit fraud, with 7% of respondents indicating 81-100% of claimants had given false information to support their claim. This perception (that 81-100 people had committed benefit fraud) has increased from 4% in 2016 (Table 2.3).

Table 2.3 – Thinking about the whole of Britain, out of every 100 people receiving benefits, how many do you think have broken the law by knowingly giving false information to support their claim?

	2016	2017	2018	2019	2022
0-20	40%	37%	35%	38%	37%
21-40	20%	20%	20%	20%	17%
41-60	17%	20%	21%	20%	18%
61-80	10%	14%	11%	11%	11%
81-100	4%	4%	5%	4%	7%
Don’t Know	8%	0%	8%	7%	9%
Refusal	0%	5%	0%	0%	1%
Unweighted bases	2,942	2,963	2,884	3,224	3,310

Source: The National Centre For Social Research, British Social Attitudes survey, 2016, 2017, 2018, 2019 & 2022

2.3. Morality of Benefit Fraud

BSA explored the acceptability of benefit fraud:

an unemployed person on benefit takes a casual job paid in cash; he does not report it to the benefit office and is £500 in pocket. Is this wrong?
an unemployed person on benefit takes a casual job paid in cash; he does not report it to the benefit office and is £3,000 in pocket. Is this wrong?
suppose someone knowingly gave false information to support their benefit claim. What would your view about this be?
a person on benefit moved in with his girlfriend and her young daughter a year ago he does not report this, and the couple are £3,000 in pocket, is this wrong?

Most respondents agreed that benefit fraud is wrong. In 2022, over half of respondents (53%) believed it was wrong (“wrong” or “seriously wrong”) if an unemployed person on benefit took a casual job and was £500 in pocket. However, this attitude decreased from 68% in 2016 and instead there has been an increase in respondents who reported that these scenarios were not wrong (“not wrong” or “a bit wrong”) from 27% in 2016, to 38% in 2022 (Table 2.4).

Table 2.4 – Consider this situation: An unemployed person on benefits takes a casual job and is paid in cash. He does not report it to the benefit office and is £500 in pocket. Do you feel that this is wrong or not wrong?

	2016	2017	2018	2019	2022
Not Wrong/ A bit wrong	27%	N/A	N/A	N/A	38%
Wrong/ Seriously wrong	68%	N/A	N/A	N/A	53%
Don’t Know	3%	N/A	N/A	N/A	7%
Refusal	2%	N/A	N/A	N/A	1%
Unweighted bases	1,618	N/A	N/A	N/A	1,646

Source: The National Centre For Social Research, British Social Attitudes survey, 2016 & 2022

Similarly in the context of a greater overpayment, 66% of respondents thought failing to disclose earnings was wrong in 2022 (“wrong” or “seriously wrong”). However, this attitude has declined since 2016 from 80%. There has also been an increase in respondents suggesting the situation is not wrong (“not wrong” or “a bit wrong”) from 16% in 2016 to 27% in 2022 (Table 2.5).

Table 2.5 – Consider this situation: An unemployed person on benefit takes a casual job and is paid in cash. He does not report it to the benefit office and is £3,000 in pocket. Do you feel this is wrong or not wrong?

	2016	2017	2018	2019	2022
Not Wrong/ a bit wrong	16%	16%	19%	20%	27%
Wrong/ Seriously wrong	80%	81%	77%	74%	66%
Don’t Know	2%	2%	3%	4%	7%
Refusal	2%	1%	1%	1%	0%
Unweighted bases	782	2,446	2,269	2,636	1,664

Source: The National Centre For Social Research, British Social Attitudes survey, 2016, 2017, 2018, 2019 & 2022

Respondents were also asked about the acceptability of a claimant intentionally providing false information for their benefit claim. In 2019 87% of respondents agreed that benefit fraud is wrong (“always” or “usually wrong”). There was evidence of a small decrease from 2016 in which 91% of respondents thought benefit fraud was wrong. Conversely, there was an increase in respondents who answered that it was “sometimes wrong”, from 7% in 2016 to 11% in 2019 (Table 2.6).

Table 2.6 – Suppose someone knowingly gave false information to support their benefit claim. What would your view about this be?

	2016	2017	2018	2019	2022
Rarely/ Never wrong	1%	1%	N/A	1%	N/A
Sometimes wrong	7%	10%	N/A	11%	N/A
Always/ Usually wrong	91%	89%	N/A	87%	N/A
Don’t Know	1%	0%	N/A	1%	N/A
Refusal	0%	N/A	N/A	0%	N/A
Unweighted bases	2,942	2,963	N/A	3,224	N/A

Source: The National Centre For Social Research, British Social Attitudes survey, 2016, 2017 & 2019

Looking at claimants failing to disclose a partner moving in and receiving a £3,000 overpayment, 76% of respondents thought this was wrong (“wrong” or “seriously wrong”) in 2019 (Table 2.7).

Table 2.7 – A person on benefit moved in with his girlfriend and her young daughter a year ago he does not report this, and the couple are £3,000 in pocket, is this wrong?

	2016	2017	2018	2019	2022
Not wrong/ A bit wrong	N/A	N/A	15%	17%	N/A
Wrong/ Seriously wrong	N/A	N/A	80%	76%	N/A
Don’t Know	N/A	N/A	2%	5%	N/A
Refusal	N/A	N/A	3%	1%	N/A
Unweighted bases	N/A	N/A	748	2,636	N/A

Source: The National Centre For Social Research, British Social Attitudes survey, 2018 & 2019

2.4. Reporting Benefit Fraud

BSA asked about the likelihood of reporting someone for benefit fraud. In 2022, 51% of respondents stated that they were unlikely (“not very likely” or “not at all likely”) to report someone for benefit fraud, whereas 46% were likely (“very likely” or “fairly likely”). The number of respondents who were unlikely to report someone for benefit fraud declined from 57% in 2016 to 51% in 2022, however this trend was mixed over that time period (Table 2.8).

Table 2.8 – Imagine that you found out that someone falsely claimed benefits amounting to £3,000. You know this person but they are not a close friend or family member. How likely or unlikely is it that you would report them?

	2016	2017	2018	2019	2022
Very/ Fairly likely	N/A	42%	N/A	39%	46%
Not Very/ Not at all likely	N/A	57%	N/A	59%	51%
Don’t Know	N/A	1%	N/A	1%	2%
Refusal	N/A	0%	N/A	0%	1%
Unweighted bases	N/A	2,963	N/A	3,224	3,310

Source: The National Centre For Social Research, British Social Attitudes survey, 2017 & 2019

2.5. Tackling Benefit Fraud

BSA asked about efforts by the Government to tackle benefit fraud. Overall, respondents thought it was unlikely (“very” or “not at all likely”) that someone who falsely claimed benefits would be caught. In 2022, 61% of respondents had this view, compared with 2016 this attitude had increased (from 54%), however this trend was mixed over that time period. Conversely, in 2022 1 in 10 respondents selected “don’t know” (Table 2.9).

Table 2.9 – How likely do you think it is that someone who falsely claims benefits will be caught?

	2016	2017	2018	2019	2022
Very/ Fairly likely	40%	44%	45%	41%	28%
Not Very/ Not at all likely	54%	52%	50%	52%	61%
Don’t Know	5%	3%	4%	6%	10%
Refusal	1%	1%	2%	1%	2%
Unweighted bases	2,400	2,446	2,269	2,636	3,310

Source: The National Centre For Social Research, British Social Attitudes survey, 2016, 2017, 2018, 2019 & 2022

Similarly, in 2022 60% of respondents did not think the government were doing enough to reduce benefit fraud. There was evidence of a small increase compared with 2017 in which 57% of respondents did not think the government were doing enough, however this trend was mixed over that time period. The proportion of respondents with the view that the government are doing ‘too much’ increased from 4% in 2017 to 8% in 2022 (Table 2.10).

Table 2.10 – Do you think the government is doing too much to reduce benefit fraud, not enough, or the right amount?

	2016	2017	2018	2019	2022
Too much	N/A	4%	7%	4%	8%
Not enough	N/A	57%	54%	52%	60%
The right amount	N/A	30%	27%	29%	24%
Don’t Know	N/A	10%	12%	14%	8%
Refusal	N/A	N/A	0%	0%	0%
Unweighted bases	N/A	2,963	2,884	3,224	3,310

Source: The National Centre For Social Research, British Social Attitudes survey, 2017, 2018, 2019 & 2022

BSA also asked about the public’s awareness of the penalties that can be applied for committing benefit fraud:

if someone is found guilty of a benefit fraud offence, certain benefits can be reduced or stopped for up to 3 years. Were you aware of this before the interview?
if someone is found guilty of benefit fraud, they are always required to pay back any extra money they were paid. Were you aware of this before the interview?
if someone commits benefit fraud, they can choose to receive a financial penalty of up to £5,000 instead of being prosecuted. Were you aware of this before the interview?
if someone is found guilty of benefit fraud, they will have a criminal record. Were you aware of this before the interview?

In 2019, 71% of respondents were not aware that if someone was found guilty of benefit fraud, certain benefits will be reduced or stopped, although a sizeable proportion (29%) were aware (Table 2.11).

Table 2.11 – If someone is found guilty of a benefit fraud offence, certain benefits can be reduced or stopped for up to 3 years. Were you aware of this before the interview?

	2016	2017	2018	2019	2022
Yes, I was aware of this	30%	25%	N/A	29%	N/A
No, I was not aware of this	70%	75%	N/A	71%	N/A
Don’t Know	0%	0%	N/A	0%	N/A
Refusal	0%	N/A	N/A	N/A	N/A
Unweighted bases	2,942	2,963	N/A	3,224	N/A

Source: The National Centre For Social Research, British Social Attitudes survey, 2016, 2017 & 2019

In 2019, 61% of respondents were aware that someone found guilty of benefit fraud must pay back any extra money they were paid. This was one of the most consistently recognised penalty for benefit fraud, out of the four penalties asked about in the survey. Since 2017, the number of respondents who were aware of this penalty, increased from 54% to 61%(Table 2.12).

Table 2.12 – If someone is found guilty of benefit fraud, they are always required to pay back any extra money they were paid. Were you aware of this before the interview?

	2016	2017	2018	2019	2022
Yes, I was aware of this	N/A	54%	N/A	61%	N/A
No, I was not aware of this	N/A	46%	N/A	39%	N/A
Don’t Know	N/A	0%	N/A	0%	N/A
Refusal	N/A	N/A	N/A	N/A	N/A
Unweighted bases	N/A	2,963	N/A	3,224	N/A

Source: The National Centre For Social Research, British Social Attitudes survey, 2017 & 2019

In 2019, 87% of respondents were not aware that if someone committed benefit fraud, they could choose to receive a financial penalty of up to £5,000 instead of being prosecuted. Although, the number of respondents who were not aware of this penalty type has decreased since 2017 (from 92% to 87%), whereas the number of respondents aware has increased (from 8% to 12%) (Table 2.13).

Table 2.13 – If someone commits benefit fraud, they can choose to receive a financial penalty of up to £5,000 instead of being prosecuted. Were you aware of this before the interview?

	2016	2017	2018	2019	2022
Yes, I was aware of this	N/A	8%	N/A	12%	N/A
No, I was not aware of this	N/A	92%	N/A	87%	N/A
Don’t Know	N/A	0%	N/A	0%	N/A
Refusal	N/A	N/A	N/A	N/A	N/A
Unweighted bases	N/A	2,963	N/A	3,224	N/A

Source: The National Centre For Social Research, British Social Attitudes survey, 2017 & 2019

In 2019, 61% of respondents were aware that someone found guilty of benefit fraud would receive a criminal record. This was one of the most consistently recognised penalty for benefit fraud, out of the four penalties asked about in the survey. Since 2017, the number of respondents who were aware of this penalty, increased from 48% to 61% (Table 2.14).

Table 2.14 – If someone is found guilty of benefit fraud, they will have a criminal record. Were you aware of this before the interview?

	2016	2017	2018	2019	2022
Yes, I was aware of this	N/A	48%	N/A	61%	N/A
No, I was not aware of this	N/A	52%	N/A	39%	N/A
Don’t Know	N/A	0%	N/A	0%	N/A
Refusal	N/A	N/A	N/A	N/A	N/A
Unweighted bases	N/A	2,963	N/A	3,224	N/A

Source: The National Centre For Social Research, British Social Attitudes survey, 2017 & 2019

There was an overall low awareness of penalties. However, since 2017 the awareness of most penalties has increased (the requirement to pay back any extra money they were paid, receiving a financial penalty instead of prosecution and receiving a criminal record) (Tables 2.11 to 2.14).

2.6. Likelihood of Committing Benefit Fraud

Following a question surrounding morality of benefit fraud (“An unemployed person on benefit takes a casual job paid in cash; he does not report it to the benefit office and is £3,000 in pocket. Is this wrong?”), participants were asked to consider the likelihood that they would commit that type of benefit fraud.

In 2019, 76% of respondents said they were unlikely (“not very likely” or “not at all likely”) to commit that type of benefit fraud. This attitude has remained steady since 2016. (Table 2.15).

Table 2.15 – And how likely that you would take a casual job and not report £3,000 to benefit office?

	2016	2017	2018	2019	2022
Very/ Fairly Likely	14%	12%	15%	15%	N/A
Not Very/ Not at all Likely	78%	82%	78%	76%	N/A
Don’t Know	5%	4%	5%	8%	N/A
Refusal	3%	1%	1%	2%	N/A
Unweighted bases	782	2,446	756	2,636	N/A

Source: The National Centre For Social Research, British Social Attitudes survey, 2016, 2017, 2018 & 2019

Question 5

3.  Attitudes to Fraud and Error in Welfare Benefits in 2022: additional questions asked

Accepted Answer

3.1. Reporting Benefit Fraud

The British Social Attitudes (BSA) survey additionally asked respondents about reporting benefit fraud in 2022.

Respondents were most likely to use the National Benefit Fraud Hotline to report benefit fraud (24%). This was followed by doing an internet search to find out (16%) and reporting it to the Department for Work and Pensions (15%) (Table 3.1).

Table 3.1 – You have said it’s likely you would report someone who has falsely claimed benefits. Who would you report this to?

How you’d report benefit fraud	Percentage
National Benefit Fraud Hotline	24%
Internet search to find out	16%
Department for Work and Pensions	15%
Benefits Agency/ Benefits Office	10%
HMRC/ Inland revenue/ Tax Office	9%
Local Authority/ Local Council	8%
Department of Social Security	7%
The police	6%
Jobcentre Plus/ Employment Service	4%
Other government department or agency	2%
Other answer	0%
Unweighted Base	2,707

Source: The National Centre For Social Research, British Social Attitudes survey, 2022.

The most common reason given in 2022 for not reporting benefit fraud was because the respondent might not have all the facts (30%). This was followed by not wanting to cause trouble for other people (13%) and not wanting to get involved (13%). There were also three further reasons given by around 10% of respondents for each: “I would worry that they would find out it was me who reported them”, 12%; “I don’t know how to report it”, 11%; and “It is not my responsibility to report benefit fraud”, 10% (Table 3.2).

Table 3.2 – You have said it is not likely that you would report someone falsely claiming benefits. What are your reasons for saying this?

Reason you’d give	Percentage
I might not have all the facts	30%
I don’t want to cause trouble for other people	13%
I don’t want to get involved	13%
I would worry that they would find out it was me who reported them	12%
I don’t know how to report it	11%
It is not my responsibility to report benefit fraud	10%
Nothing would happen as a result of me reporting it	4%
I don’t care/ it doesn’t bother me	4%
Tax fraud/ corporate fraud is a bigger issue	1%
Other	1%
Unweighted Base	3,949

Source: The National Centre For Social Research, British Social Attitudes survey, 2022

Question 6

4.  Attitudes to Fraud and Error in Welfare Benefits in 2018: additional questions asked

Accepted Answer

4.1. Media Coverage of Benefit Fraud

The British Social Attitudes (BSA) survey asked additional questions about the coverage of benefit fraud in the media in 2018.

Respondents largely reported seeing or hearing on TV, in the news or online about benefit fraud in the last 12 months: “a great deal or quite a lot” (36%); or “some” (34%). Conversely a similar level reported seeing “not very much or nothing at all” (29%) (Table 4.1).

Table 4.1 – How much have you seen or heard on TV, in the news or online about benefit fraud in the last 12 months?

How you’d report benefit fraud	Percentage
National Benefit Fraud Hotline	24%
Internet search to find out	16%
Department for Work and Pensions	15%
Benefits Agency/ Benefits Office	10%
HMRC/ Inland revenue/ Tax Office	9%
Local Authority/ Local Council	8%
Department of Social Security	7%
The police	6%
Jobcentre Plus/ Employment Service	4%
Other government department or agency	2%
Other answer	0%
Unweighted Base	2,707

Source: The National Centre For Social Research, British Social Attitudes survey, 2018

Nearly half, 46% of respondents agreed that this was the right amount of coverage about benefit fraud in 2018, however, 14% stated it was too much coverage and 33% that it was not enough coverage (Table 4.2).

Table 4.2 – Is there too much, not enough or the right amount of coverage about benefit fraud on television, in the news or online?

Amount of coverage	Percentage
Too much	14%
Not enough	33%
The right amount	46%
Don’t Know	7%
Refusal	0%
Unweighted Base	2,884

Source: The National Centre For Social Research, British Social Attitudes survey, 2018

4.2. Morality of Benefit Fraud

BSA asked about how acceptable benefit fraud is in 2018.

93% of respondents agreed that it is unacceptable (“rarely” or “never acceptable”) for someone to knowingly give false information to support their benefit claim. Conversely, 1% thought it was “always or often” acceptable and 6% thought it was “sometimes” acceptable (Table 4.3).

Table 4.3 – How acceptable do you think it is for someone to knowingly give false information to support their benefit claim?

Level of acceptance	Percentage
Always/ often acceptable	1%
Sometimes acceptable	6%
Rarely/ never acceptable	93%
Don’t Know	1%
Unweighted Base	1,455

Source: The National Centre For Social Research, British Social Attitudes survey, 2018

This view of acceptability, on giving false information to support a benefit claim, was lower if the question wording was changed to indicate that they were struggling to pay their rent or heating bills.

77% thought it was “rarely or never” acceptable if someone was struggling to pay their rent or heating bills to give false information.

With a greater proportion thinking it was “sometimes” acceptable in this scenario (18%), or “always or often” acceptable (4%) (Table 4.4).

Table 4.4 – If someone is struggling to pay their rent or heating bills, how acceptable do you think it is for them to knowingly give false information to support their benefit claim?

Level of acceptance	Percentage
Always/ often acceptable	4%
Sometimes acceptable	18%
Rarely/ never acceptable	77%
Don’t Know	1%
Unweighted Base	1,429

Source: The National Centre For Social Research, British Social Attitudes survey, 2018

Regarding the acceptability of disability benefit fraud, 83% of respondents stated it was wrong (“wrong” or “seriously wrong”) that a person had been receiving extra benefits since a back injury stopped them working but is working and is £3,000 in pocket (Table 4.5).

Table 4.5 – A person has been receiving extra benefits since a back injury stopped him working; he does work but does not tell the benefit office and is £3,000 in pocket. Is this wrong?

Right or wrong	Percentage
Not wrong/ a bit wrong	12%
Wrong/ Seriously wrong	83%
Don’t Know	1%
Refusal	3%
Unweighted Base	765

Source: The National Centre For Social Research, British Social Attitudes survey, 2018

BSA asked about the government data sharing for the advantage of tackling benefit fraud in 2018. 72% of respondents were comfortable (“very comfortable” or “fairly comfortable”) with the government comparing benefit claims with tax records to find people who are claiming too much while working. A further 15% were neutral about data sharing. However, there were over 1 in 10 respondents (12%) who were “fairly or very” uncomfortable with data sharing (Table 4.6).

Table 4.6 – The government currently compares information from people’s benefit claims with their tax records, in order to find people who are claiming too much in benefits while they are working. How comfortable or uncomfortable are you with the government doing this?

Level of comfort	Percentage
Very/ fairly comfortable	72%
Neither comfortable nor uncomfortable	15%
Fairly/ very uncomfortable	12%
Don’t Know	2%
Unweighted Base	2,884

Source: The National Centre For Social Research, British Social Attitudes survey, 2018

A large proportion of respondents (63%) were not aware that when someone who claims benefits goes into hospital, some of their benefit payments may be reduced temporarily while they are in hospital (Table 4.7).

Table 4.7 – When someone who claims benefits goes into hospital, some of their benefit payments may be reduced temporarily while they are in hospital. Were you aware of this before this interview?

Yes/No	Percentage
Yes, I was aware of this	37%
No, I was not aware of this	63%
Don’t Know	0%
Unweighted bases	2,884

Source: The National Centre For Social Research, British Social Attitudes survey, 2018

Over half of respondents in 2018 (56%) were comfortable (“very comfortable” or “fairly comfortable”) with the government using hospital admissions records to make sure people on benefits were paid correctly if they need to go into hospital. However, a sizeable proportion were not comfortable with this form of data sharing (29%) (Table 4.8).

Table 4.8 – Imagine if the government were to use hospital admissions records to make sure people on benefits are paid correctly if they need to go into hospital. How comfortable or uncomfortable are you with the government doing this?

Level of Comfort	Percentage
Very/ fairly comfortable	56%
Neither comfortable nor uncomfortable	13%
Fairly/ very uncomfortable	29%
Don’t Know	1%
Unweighted Base	2,884

Source: The National Centre For Social Research, British Social Attitudes survey, 2018

Question 7

5.  Attitudes to Fraud and Error in Welfare Benefits in 2017: additional questions asked

Accepted Answer

5.1. Penalties for Benefit Fraud

The British Social Attitudes (BSA) survey asked about penalties for benefit fraud in 2017. The most common penalties recognised by respondents were: firstly paying back overpayment of benefits (30%); followed by a fine (16%); community service (13%); stopping their benefit payments (12%); and criminal record (10%) (Table 5.1).

Table 5.1 – Now thinking about someone who is caught claiming £3,000 of benefits they are not entitled to. It is the first time they have done this. Which, if any, of the punishments in this list do you think they would receive?

Type of punishment	Percentage
Paying back overpayment of benefits	30%
A fine	16%
Community Service	13%
Stopping their benefits payments	12%
Criminal Record	10%
Reducing their benefit payments	9%
Having their name made public (e.g., in local press)	5%
Prison	2%
Having their home or possessions taken away	1%
No punishment at all	1%
Other	0%
Unweighted Base	2,926

Source: The National Centre For Social Research, British Social Attitudes survey, 2017

Question 8

6.  Attitudes to Fraud and Error in welfare Benefits in 2016: additional questions asked

Accepted Answer

6.1. Media Coverage of Benefit Fraud

The British Social Attitudes (BSA) survey asked about benefit fraud in the media in 2016. 31% of respondents only remembered seeing one to two media articles on benefit fraud in the last six months. Conversely, 31% respondents had seen between 3-10 articles and 11% had seen more than 10 (Table 6.1).