WIREDGOV

Executive summary

This report examines ethnic differences in pension participation in the UK following the nationwide roll-out of automatic enrolment, the drivers of those gaps, and the consequences for wealth accumulation. It draws on an academic research paper (Cribb, O’Brien and Sturrock, 2025) published alongside this report which provides a fuller examination of these issues.

Introduction 

There is rising interest in wealth inequalities between different groups in society, including between people of different ethnic backgrounds. The UK is an ethnically diverse country, with around 9% of the population of England and Wales from an Asian ethnic group and 4% from a Black group according to the 2021 Census1.  Average wealth holdings differ substantially by ethnicity in the UK; for example, median wealth for households with a White British household head is over £300,000, compared with under £70,000 for those with a Black African or Bangladeshi household head (Office for National Statistics, 2020). These wealth inequalities are important because of the ability of private wealth to insure families against adverse shocks, to provide an income in retirement and to be passed down to future generations. 

However, understanding ethnic inequalities in wealth accumulation is complex and differences between groups could occur purely because of age differences, how long people have lived in the UK, or because people with lower lifetime incomes may appropriately accumulate lower levels of retirement wealth, with the state pension providing much better replacement rates for lower earners than for higher earners. It is therefore important to examine specific mechanisms that may drive differences in wealth accumulation and how they interact with ethnicity.

This short report summarises new findings on ethnic differences in private pension participation in the UK, the drivers of those gaps and the consequences for wealth accumulation. This report draws heavily on a research paper published alongside this report (Cribb, O’Brien and Sturrock, 2025), which builds on a growing academic literature in this area (Derenoncourt et al., 2024; Choukhmane et al., 2024).

Importantly, most employees in the UK (those aged 22 to state pension age and earning over £10,000 per year) are automatically enrolled by their employer into a workplace pension scheme, though they can choose to leave the scheme (opt out) at any point if they wish. In the public sector, employees are generally auto-enrolled into a defined benefit pension plan, whereas in the private sector it is generally a defined contribution arrangement. For those who participate, there are minimum contributions (8% of ‘qualifying’ earnings, with at least 3% coming from the employer), though average contribution rates are higher, particularly in the public sector, where employer contributions alone are typically worth at least 23% of salary (Cribb, Dominguez and McKendrick, 2024). 

Previous research has shown that automatic enrolment has both boosted pension participation substantially, to around 90% of eligible employees, and significantly reduced inequalities in pension saving by earnings and age (Cribb and Emmerson, 2020). This summary report shows, however, that even after automatic enrolment, there are still big differences in pension participation rates between ethnic groups, and these have important implications for incomes in retirement. 

How and why do private pension participation rates differ by ethnicity in the UK?

Figure 1 shows the key differences in pension participation along the lines of ethnicity for employees in the UK. Looking only at employees who are eligible for automatic enrolment (AE), it shows that, after being automatically enrolled, 10% of White employees opt out of their workplace pension plan, and although most other ethnicities have similar opt-out rates2, this rises to 16% for employees from a Pakistani background and 24% for employees from a Bangladeshi background (or 18% when looking at Pakistanis and Bangladeshis together). This is a striking difference compared with other ethnic groups. The opt-out rate among all employees who are matched to the Census and who are not of Pakistani or Bangladeshi origin is 9.7%.

One potential reason for these differences could be the characteristics of the individual, their job and their employer. For example, lower-paid employees are more likely to opt out than higher-paid ones (Cribb and Emmerson, 2020). Figure 2 therefore shows how controlling for different characteristics affects gaps in opt-out rates between different ethnic minority groups of employees compared with the White majority. Positive numbers here therefore imply higher opt-out rates for the ethnic minority group. The green bars show the differences in opt-out rates not controlling for any characteristics, and reflect the differences shown in Figure 1. The blue bars subsequently control for age, earnings, education (interacted with age), region, sex, migrant status, part-time/full-time worker, occupation, and the sector, industry and size of the specific employer worked for. The gaps between White employees and Pakistani and Bangladeshi employees are barely changed compared with the raw differences. Therefore even examining similar employees working for the same employer, compared with White employees, Pakistanis are still 5 percentage points more likely to opt out of saving in a pension and Bangladeshis are 13 percentage points more likely.

The yellow bars in Figure 2 additionally control for each individual’s religion. Controlling for religion eliminates the gap in pension participation for Pakistanis and reduces it to 9 percentage points for Bangladeshis. Given that around 90% of Pakistanis and Bangladeshis are Muslim, this points to religion, and Islam specifically, potentially being an important determinant of pension participation behaviour.

Indeed, we find a range of evidence that points towards religious beliefs and/or norms in Islam being important in driving higher pension opt-out rates. Importantly, Islamic teaching generally bans particular investments: specifically, no interest income may be received and investments may not be in ‘unethical’ industries such as alcohol or tobacco.

Figure 3 shows a key piece of evidence: Muslims who report that their religious beliefs make a great deal of difference to their life are particularly likely to opt out of workplace pension participation. Controlling for their individual (e.g. age, sex, education) and job (e.g. earnings, occupation) characteristics, Muslims who report that religion makes a great difference to their life (‘more religious’ Muslims) are 12 percentage points more likely to opt out than non-religious employees, whereas ‘less religious’ Muslims are 5 percentage points more likely to opt out than non-religious employees. This means that 20% of ‘more religious’ Muslims opt out of workplace pension participation, compared with 12% of ‘less religious’ Muslims and 6% of Christian or not religious employees. Muslims from both Pakistani and Bangladeshi backgrounds, as well as from other ethnic backgrounds (such as Indian or Black African), are more likely than other groups to opt out of pension participation.

Figure 4 shows another piece of evidence that points to Islamic teaching on investments being important in driving saving behaviour: over a quarter of Muslim employees (from either Pakistani/Bangladeshi backgrounds or other backgrounds) report that their religious beliefs affect their saving and investment decisions a lot, with almost another quarter reporting that their religious beliefs affect these decisions a little, both of which are far higher than for other religions. Consistent with this, we find that Muslim employees from Pakistani/Bangladeshi backgrounds (and Muslims more generally) are significantly less likely to hold assets in interest-bearing accounts than White employees. In addition, we find that these gaps are not likely to be driven by differences in beliefs over the rate of return on different forms of saving. Employees from Pakistani/Bangladeshi backgrounds have similar beliefs to people of other ethnicities over whether private pensions are the safest or highest-return form of saving for retirement.

Finally, we show how important these patterns in pension participation are for wealth accumulation. We do this by showing how much higher incomes in retirement could be for individuals who participate in private pensions over the course of a career compared with those who opt out. Here we assume that employees who opt out of pension participation make equivalent individual contributions into an ISA for use in retirement, and they can achieve the same rate of return (net of fees) in that ISA as in a private pension.3

Figure 5 shows that participating in a pension boosts retirement income (including the state pension) by 60% in total for an average-earning Pakistani/Bangladeshi worker with average pension contributions. This increase is worth around £16,500 in cash terms. Most of it comes from employer pension contributions, which opting-out employees miss out on, but in addition people participating in private pension schemes benefit from tax relief on employee and employer pension contributions. This highlights how important persistently opting out of private pension saving can be for retirement wealth accumulation.

Conclusion and policy implications

Our research highlights a substantial disparity in pension participation among ethnic groups in the UK workforce, with employees from Pakistani and Bangladeshi backgrounds almost twice as likely to opt out of workplace pensions (18%) compared with White employees (10%). These gaps persist even when comparing people with similar individual and job characteristics and working for the same employer. Most Pakistani and Bangladeshi employees are Muslims and there is a broad set of evidence suggesting that religious beliefs and norms in Islam play a central role in driving lower rates of private pension participation. This is despite most public service pension schemes being understood to be Sharia-compliant4, and there generally being Sharia-compliant investment options for people enrolled into defined contribution pension schemes. Moreover, we estimate that saving in a workplace pension could increase retirement income by 60% for Pakistani and Bangladeshi employees compared with saving in an ISA, due to employer pension contributions and the tax advantages of private pensions.

What are the policy implications of these results? First, these findings suggest there is a case for increasing awareness amongst Muslims in the public sector regarding the Sharia-compliance of public sector defined benefit pension schemes. Similarly, for employees offered private sector defined contribution pension arrangements, policymakers concerned by these gaps could address them by ensuring all have access to Sharia-compliant investment funds, raising awareness of these funds and reducing the hassle of switching to them.

More broadly, the findings reiterate the obvious point that the only people who benefit from the employer contributions and tax advantages of private pension saving are those who participate in private pensions. Pakistani and Bangladeshi employees are therefore more likely to miss out. This means that a move towards compulsory provision of (at least a low level) of employer pension contributions, irrespective of whether the employee contributes, as suggested by the IFS Pensions Review(Cribb et al., 2024), would disproportionately benefit Pakistani and Bangladeshi employees and would likely act to slightly reduce some ethnic gaps in wealth accumulation.

In contrast, these findings highlight how certain groups would be differentially affected by proposals to increase minimum employer pension contributions. For example, there is often discussion about raising minimum employer contributions to 6% of all earnings (e.g. Broome, Mulheirn and Pittaway, 2024). Employees who opt out would – at best – not benefit from these proposals. At worst, higher minimum pension contributions for those participating could push down pay (as is likely; see Cribb et al. (2024) for a review of the evidence). This would leave those who opt out (disproportionately Pakistani and Bangladeshi employees) unambiguously worse off – they would have lower pay and no boost in their provision for retirement.

Finally, given the rising fraction of the population that is from a Bangladeshi or Pakistani background (or indeed is Muslim), the potential gap in wealth accumulation generated by differing pension participation decisions and the salience of some of these gaps, there is a good case for more specific research into this issue and the potential for trialling a range of options to encourage appropriate pension participation for Muslims in products that are compliant with their religious beliefs.

References

Broome, M., Mulheirn, I. and Pittaway, S., 2024. Precautionary tales: tackling the problem of low saving among UK households. Resolution Foundation, https://www.resolutionfoundation.org/publications/precautionary-tales/.

Choukhmane, T., Colmenares, J., O’Dea, C., Rothbaum, J. L. and Schmidt, L. D. W., 2024. Who benefits from retirement saving incentives in the U.S.? Evidence on gaps in retirement wealth accumulation by race and parental income. National Bureau of Economic Research, Working Paper 32843, https://www.nber.org/papers/w32843.

Cribb, J., Dominguez, M. and McKendrick, A., 2024. Pressures on public sector pay. Institute for Fiscal Studies, Report, https://ifs.org.uk/publications/pressures-public-sector-pay.

Cribb, J. and Emmerson, C., 2020. What happens to workplace pension saving when employers are obliged to enrol employees automatically? International Tax and Public Finance, 27, 664–93, https://doi.org/10.1007/s10797-019-09565-6.

Cribb, J., Emmerson, C., Johnson, P., O’Brien, L. and Sturrock, D., 2024. Policies to improve employees’ retirement resources. Institute for Fiscal Studies, Pensions Review Report, https://ifs.org.uk/publications/policies-improve-employees-retirement-resources.

Cribb, J., O’Brien, L. and Sturrock, D., 2025. Ethnic differences in retirement wealth accumulation in the UK. Institute for Fiscal Studies, Working Paper 25/02, https://ifs.org.uk/publications/ethnic-differences-retirement-wealth-accumulation-uk.

Derenoncourt, E., Kim, C. H., Kuhn, M. and Schularick, M., 2024. Wealth of two nations: the U.S. racial wealth gap, 1860–2020. Quarterly Journal of Economics, 139, 693–750, https://doi.org/10.1093/qje/qjad044.

Office for National Statistics, 2020. Household wealth by ethnicity, Great Britian: April 2016 to March 2018. https://www.ons.gov.uk/peoplepopulationandcommunity/personalandhouseholdfinances/incomeandwealth/articles/householdwealthbyethnicitygreatbritain/april2016tomarch2018.

Data

Office for National Statistics, released 03 July 2023, ONS SRS Metadata Catalogue, dataset, Annual Survey of Hours and Earnings linked to 2011 Census - England and Wales, https://doi.org/10.57906/80f7-te97.

University of Essex, Institute for Social and Economic Research. (2024). Understanding Society. [data series]. 13th Release. UK Data Service. SN: 2000053, DOI: http://doi.org/10.5255/UKDA-Series-2000053.

Related content

Employees of Pakistani or Bangladeshi ethnic origin almost twice as likely to opt out of workplace pensions as other employees

1. 1 Just under 4% of the population of England and Wales identified their ethnic group as ‘Bangladeshi’ or ‘Pakistani’.
2. 2 Opt-out rates are defined here as ‘1 – pension participation’, so employees can have left their pension scheme either shortly after being automatically enrolled (formally termed by the Department for Work and Pensions as ‘opting out’) or at a later point (formally termed by the Department for Work and Pensions as ‘cessation’).