Retirement saving and the self-employed

There’s been a welcome renewed focus recently on how to address the lack of long-term saving among those outside of traditional employment: the self-employed, contingent workers and those employed through the ‘gig economy’. The success of auto enrolment so far has been great to watch but it has shone a light on the remaining challenge faced by those who don’t have an employer to ‘default’ them into saving or match their contributions. And, the scale of this issue is increasing as the number of people in these groups continues to grow.

We’ve been looking at this challenge with i2 Media and working closely with the Royal Society for the Encouragement of Arts, Manufactures and Commerce (RSA) and Britain Thinks, both of whom have published reports on this topic.1 We also held a roundtable discussion with them and a number of experts from across sectors, which our own report, published today, summarises: Retirement saving and the self-employed (PDF)

Much of the debate has focused on legislative or regulatory changes to try to mimic auto enrolment for these groups. Ideas include using the tax self-assessment system to automatically deduct contributions and, as the RSA suggests in their report, looking at ways to adjust tax relief to reflect a contribution match. All of these suggestions have merit and are worth considering.

Our interest, though, lies more in understanding what interventions might be possible without, or ahead of, policy change. We’re not convinced there really is an equivalent of auto enrolment for these workers that would be as effective at getting them to start saving. True auto enrolment involves a default applied to the worker by someone else, such as an employer. This can’t be directly replicated in self-assessment, where ultimately it’s still the saver taking action. Annualised models for contributions also risk increasing the perceived cost of contributing, by rolling up a year’s worth of contributions into a single number.

This probably means it will be difficult to reach comparable levels of participation among contingent workers to those we’ve seen among the automatically enrolled population. But there’s still cause for optimism. In terms of the initial take-up decision, we’ve seen great work by academics at the University of Toronto, working with Ideas42 and the Mexican Government, to drive voluntary retirement saving there.2 They focused on the impact of emphasising the ways a worker’s family might benefit from them saving, based on the insight that many felt that it was selfish to prioritise saving for the future over current family needs. Another excellent recent study by Shlomo Benartzi and Hal Hershfield found that presenting contribution values as daily, rather than monthly, amounts could considerably increase voluntary take up, especially among those on lower incomes.3

While the pure inertia of auto enrolment may not be replicable for these groups, we’re also interested in whether other elements of the workplace experience of saving can be replicated, in particular the ‘set and forget’ dimension of payroll deduction. So, for example, workers paid through gig-economy platforms could be offered the option to pre-commit a percentage of each fare or fee to a savings product. Self-employed workers who invoice electronically could also be offered such a feature.

Finally, we’re curious as to whether the structure of products offered to these groups might affect take-up. For example, attitudes to risk may differ, and this may lead to preferences for more guaranteed products. Equally, differing liquidity needs may make the illiquid nature of a traditional pension less appealing.

All of these are just hypotheses or areas to explore, so over the next few months we’ll be working with academic and industry partners to try to implement research trials testing some of these ideas. We think a lot could be done to increase long-term saving among the self-employed without any policy or regulatory intervention, and we look forward to testing out what works and what doesn’t for these groups.

Will Sandbrook, executive director of Nest Insight

Dellot, B and Wallace-Stephens, F. (2018) Venturing to Retire Boosting the long-term savings and retirement security of the self-employed. Available here.
Britain Thinks (2018) The Self-Employed and Retirement Planning.

Fertig, A., Lefkowitz, J., and Fishbane, A. (2015) Using Behavioral Science to Increase Retirement Savings. Available here.

3 Hershfield, H.E., Shu, S. and Benartzi, S. (2018) Temporal Reframing and Savings: A Field Experiment