How much can DC pension savers expect to retire from their current pool? This turns out to be a remarkably tricky question to answer – and one where the rules are about to be rewritten.
As a result of the proposed changes, millions of people in the UK could find that from October 2023 their pension statements show significantly different projections from previous statements, particularly for the many years away from retirement.
Since automatic enrollment for occupational retirement began in 2012, more than 10 million people have started saving for retirement for the first time.
The vast majority of them have been enrolled in a defined contribution (DC) pension plan – an investment product designed to give them a sum of money to support themselves in retirement. Several million more already held CD savings through personal pensions or stakeholder pensions or similar vehicles.
As a retirement saver, you can reasonably assume that pension companies and pension plans have strict rules about how they project your current retirement pot into estimated retirement income. But these rules, currently under the auspices of the Financial Reporting Council (FRC), give pension providers remarkable leeway in how they go about pension projections.
In particular, providers can vary widely in the rate of return they assume members can expect for different types of investment.
These differences are clearly illustrated in a graph published by the FRC in 2020. This one is based on the anonymized responses of 17 insurance companies, wealth managers and consultancy firms which between them issue well over 20 million pension statements per year.
The top line shows the assumed rate of return on equity investment, with growth rates ranging from around 4-7%. Assumed corporate bond yields range from 1 to 4 percent, gilts from less than 0.5 percent to about 4 percent, and cash from about 0.5 to 1.5 percent.
Even over a year, this would cause significant differences, but the cumulative effect of this variation in assumptions could have a dramatic effect on the size of a projected retirement pool.
To give a simple example, consider a £50,000 retirement pot held by someone who is 20 years from retirement and invested entirely in equities. If the provider assumes 4% growth per year, this would generate a forecast pot of just under £110,000. But with 7% annual growth, the pot should be worth just over £193,000, or around three-quarters more for the same current retirement pot.
Variations like this are worrying enough as they stand, but with the advent of the proposed ‘dashboard’ for pensions, the case for change becomes overwhelming. On the dashboard, savers will be able to see all their pensions in one place. Without reform, they will see different pensions projected in totally different and inconsistent ways and will struggle to make sense of the numbers.
In response, the FRC concluded a consultation this week in which it proposes a standardization of investment growth assumptions. These changes, if implemented, would apply to all statements issued after October 2023, regardless of progress made at this stage on the pension dashboards project.
At first glance, the FRC’s desire for uniformity makes perfect sense. It will be hard enough for savers to understand the information they see on a retirement dashboard without having huge inconsistencies in the underlying investment return assumptions between different retirement providers.
However, there are serious concerns about how the FRC plans to implement these changes. In order to standardize growth assumptions, an obvious approach would be to specify for each major asset class the rate of return to be used by providers. These assumptions may be reviewed periodically as market conditions change.
Unfortunately, the FRC did not take this route.
Instead, he proposes to model future investment growth based on the volatility of returns in the recent past. Based on the principle that higher risk investments tend to be associated with higher rates of return, the FRC argues that if an investment fund has been volatile in value in recent years, it is likely to be a high-risk fund and can therefore be expected to generate high rates of return in the future.
Conversely, an investment fund that has been relatively stable in value over the past several years would be considered “low risk” and therefore “low return” when projecting future returns.
Although there is some logic in this approach, it could produce perverse results. Had it been in effect five years ago, for example, many equity fund values would have been seen as relatively stable, while some government bond yields would have been relatively volatile.
Applying FRC logic, this could mean that where a saver has invested heavily in stocks, the provider may have to project a low rate of growth; when the saver has invested heavily in bonds, the provider may have to assume a high growth rate. This is the exact opposite of the approach currently taken (as the chart shows) and also the opposite of what we would expect to see in practice.
Any switch to a new pension statement database is likely to create upheavals and will raise many questions from savers who could suddenly notice major changes on their pension statements. For some savers, it could be even more dramatic if it coincided with a sharp change in the current value of their investments during a period of market turbulence.
If standardization is to come, it will have to be done on a relatively simple basis that savers can understand. There is a real risk that if the FRC’s proposed approach is adopted, the new survey figures will make little more sense than the old ones.
Sir Steve Webb and David Everett, who contributed to this article, are partners at consultancy LCP.