When we published Balancing today and tomorrow in June, we set out a pragmatic approach to building an emergency savings element into the auto enrolment framework. In that blueprint, we highlighted the tensions that exist between different pension reform objectives, especially the challenge of increasing pension saving while ensuring the system remains accessible and beneficial for lower earners.
Since publication, we’ve had many thoughtful conversations with policymakers, employers, providers and industry experts. Over the coming weeks we will delve deeper into some of the issues and questions that have arisen through those conversations.
In this blog, we discuss the question of how and why our thinking shifted away from the need for a savings cap or automatic rollover of savings into pensions as a mandatory element of the approach.
The idea behind sidecar savings is compelling. Bringing emergency and retirement saving together in a single workplace solution addresses two problems at once: people lacking financial resilience today and undersaving for a dignified retirement tomorrow.1 The model has attracted interest among pensions policymakers for several years, and the Centre for Inclusive Money at Nest, formerly Nest Insight, has shown how it could work in practice.
In our ‘Jars’ trials, employees saved into an instant-access savings account, or jar, through payroll alongside their regular workplace pension contributions.2 Once they reached a pre-set savings target, any further contributions were automatically redirected into their pension as additional voluntary contributions. Employees could withdraw from the savings jar for any reason. If their savings account balance dropped below the target, payroll contributions were redirected back to rebuilding it.
But in Balancing today and tomorrow, we set out a different approach – Pension-Adjacent Emergency Savings Accounts (PAESAs) – with no savings cap or automatic rollover. Instead, we propose keeping the emergency savings account separate from pensions, with any connection between the two ‘pots’ to be made through targeted, and tailored nudges at appropriate moments. This approach maintains the aims of limiting over-exposure to cash and supporting people to save more for retirement saving when they can, while offering some other advantages over a ‘hard’ savings cap and automatic rollover design.
A model with no automatic rollover would be more flexible for savers
To enable an automatic rollover a default threshold at which emergency savings stop and pension saving begins would need to be chosen. But the amount someone needs for a meaningful emergency buffer depends on many factors such as their housing tenure, household size, health and caring responsibilities.3 A one-size-fits-all ceiling would inevitably be too high for some and too low for others. Furthermore, any threshold would also require regular review and uprating over time to remain appropriate, adding further administrative complexity.
Even the destination of a rollover should be given careful consideration. For some employees, extra money above an emergency buffer may be best placed in a pension. For others, it may not. Someone saving for a first home, for example, may be poorly served by having additional savings moved into a product they cannot access until later life. An ISA or another medium-term savings product may be a better fit.
Technology has advanced since the Jars trial, creating more flexible and targeted ways to encourage additional pension saving. Targeted Support, open banking and existing Consumer Duty requirements could enable a more personalised ‘soft rollover’. This could prompt people to move savings above a suitable emergency buffer into a pension or another appropriate product, or to redirect some or all future cash savings contributions into their pension. Similar nudges could be used at key moments, such as pay rises or job changes.
A model with no automatic rollover would be simpler for employers to implement
A critical objective for us was to ensure that PAESAs could be implemented by all those across the delivery ecosystem. On paper, an automatic rollover idea is elegant. But our work with partners, and the lessons from our trials, suggests that it would add complexity and administrative burden, without necessarily enhancing pension outcomes significantly.
The main complexity is that an automatic rollover model needs an actor to know several things at once: the employee’s pay, their contribution amount, the balance in their emergency savings account, and the right pension provider to receive any rollover contributions. In the Jars trial, Salary Finance, the employee benefits provider which offered Jars, played that coordinating role. At national scale, there would need to be an actor or actors to do the same reliably for millions of workers and employers.
While payroll might look like the obvious candidate given that they already manage salary processing and pension contributions, employers’ payroll systems differ widely in sophistication. A rollover mechanism could be more difficult to implement for employers with more basic systems, for instance, only half (53%) of micro businesses used payroll software in 2024.4
In addition, savings and pensions are, for the most part, regulated separately – with pensions overseen by The Pensions Regulator (TPR) and savings regulated by the Financial Conduct Authority (FCA). Aligning regulatory oversight across the two would be complex and could take considerable time to agree and implement.
Different pension tax relief mechanisms – Relief at Source and Net Pay – could add an additional complication for payroll and pension providers in facilitating the rollover.
For these reasons, keeping the two pots separate, rather than hard wiring them together, makes the model easier for employers to implement by building on existing workplace savings infrastructure. It would also enable employers that already provide payroll savings schemes alongside their workplace pension to more easily make adjustments using their existing set-up to become compliant.
None of this means rollovers should necessarily be ruled out forever. Payroll technology, open banking and better data sharing may eventually make more personalised and automatic rollovers feasible. Some providers may already be able to build this feature in. But through extensive research and stakeholder dialogue, we have come to the conclusion that this feature is not essential to a sidecar design, and that there are genuine advantages to a ‘softer’ more flexible approach to nudging people to save more for retirement when they are ready to.
Molly Broome – Research and Policy Lead
To find out more about our work on workplace emergency savings, visit:
https://www.inclusivemoney.org.uk/research-projects/workplace-emergency-savings
To stay up to date with our latest research and events, sign up to our mailing list
1 Nest Insight (2023). Workplace sidecar saving in action.
2 The Jars trial was conducted with five employers – BT, ITV, StepChange, Timpson and the University of Glasgow.
3 Resolution Foundation (2025). Money on my mind: Understanding the savings, debt and financial resilience of low-to-middle income families.
4 Department for Business and Trade (September 2025), Longitudinal Small Business Survey 2024: SME employers (businesses with 1 to 249 employees).