The self-employment pension gap: storing up trouble for the future?
The proportion of self-employed adults who are paying in to a pension each year has been declining, falling to just 17 per cent in 2013/14. In recent research for Citizens Advice we looked at why this might be the case. In addition to low pension contribution levels, we found that the pension pots of self-employed households were also worth less than employee pension pots. Though some people could be compensating by investing in property, it is not clear that any accumulated wealth will be sufficient to offer security in retirement, with implications for the future living standards of the self-employed.
Over the last fifteen years, the number of self-employed people contributing to a pension has been in decline even as the number of self-employed has been increasing. The Family Resources Survey puts the proportion of self-employed people who are contributing to pensions at just 17% in 2013/14, compared to 52% of employees that were paying in, and ten years earlier the proportion was closer to a third (36 per cent). This trend is confirmed if you look at the records submitted by pension providers to HMRC. They suggest that 1.1 million self-employed people were contributing to a personal pension in 2003/04 but the number has more than halved since, falling to just 450,000 people in 2013/14.
Why might this be happening? Are low pension contribution rates linked to low incomes? The next graph shows that as household income rises the proportion of the self-employed that are actively contributing to a pension remains low, suggesting that low incomes are not the only factor in explaining low pension contribution levels. In fact, the proportion of self-employed adults that are paying in to a pension never rises above 24%, even when they have a combined household income of £1,000 or more per week and so, presumably, have more disposable income. In contrast, there is a clearer ‘income gradient’ amongst employees: the proportion actively contributing to a pension reaches 64% amongst the richest households.
Graph1: Proportion of adults contributing to a pension by weekly household income and employment status, 2013/14
Source: Family Resources Survey, DWP. Estimates for adults in households with an income of less than £100 are not presented due to small sample sizes.
Low pension saving rates are accompanied by lower value pension pots where the self-employed do have pension savings . For those close to retirement (aged between 55 and 64), median pension savings among self-employed households were valued at around £125,000 in 2010/12 (see Graph 2), which compares unfavourably to the average household pot for employees (£235,000).
Graph 2: Value of household pension savings by age and employment status of household head, 2010/12
Source: Wealth and Assets Survey
Given these low pension savings, and low contribution levels, it may be that the self-employed are preparing for retirement in other ways, perhaps by building up property wealth instead. Indeed, the self-employed are more likely to have this kind of wealth and the median value amongst self-employed households is higher (at £225,000 for households headed by someone aged between 45 and 54, see graph 3) than for employee households (with median wealth of £145,000 for this age group).
This offers some support to the hypothesis that property investments may serve as a resource for the self-employed in retirement. But it should not be assumed that this solves the problem of low pension contribution levels: after all, there is no guarantee that property can substitute for a fixed, cash income in old age. Instead, gambling on rising house prices to make up for a lack of pension saving may turn out to be a risky strategy. If your business fails then business debts could easily eat up any property wealth and there is also no guarantee that mortgage interest rates will continue at their historic, low levels.
Graph 3: Household property wealth 2010/12
Source: Wealth and Assets Survey. Data for employee and self-employed households with property wealth greater than zero, values are rounded
All of this has implications for individuals, in terms of income security in old age, but also for policy makers. Even a short amount of time spent not contributing to a pension could have a long-term impact on living standards in retirement. In addition, the low value of pension savings pots for those near retirement has real implications for government spending in the years to come. Those who do not save enough will have little choice but to look to the state to provide them with a minimum income in retirement. In this context, nudging people to pay into pensions – perhaps by including an auto-enrolment option on self-assessment forms as Citizens Advice suggests – could be an important first step in boosting pension coverage.