CDC schemes are likely to use the annual pension increase as the lever to ensure the scheme stays in balance, with assets always sufficient to meet target-accrued pensions. The pension increase percentage each year is calibrated such that, if you assume the same increase will be given to every member every year in the future, the best estimate 'actuarial value' of all accrued pensions is equal to the value of the scheme’s assets. This calibration takes place every year and therefore pension increases vary over time and are not guaranteed.

Across the generations

Everyone in the scheme, irrespective of age (from new joiners aged 20 to pensioners in their 90s), gets the same percentage increase to their accrued pension. It’s this element of the scheme that can result in significant investment risk-sharing across generations. What does this mean? Younger members take on more investment risk and older members take on less. But how?

Firstly, to understand investment risk-sharing, we should view a member’s benefits as an accrued pension with associated increases, but also look at how the present value ('actuarial value') of these benefits changes over time. Through this lens, we discover that a change to the pension increase award (and hence the assumptions about future increases) has a different impact on the actuarial value of benefits depending on the member’s age. For example, a 1% fall in the pension increase award (and future projected increases) for a 65-year-old will reduce the 'actuarial value' of their target benefits by around 12%, whereas the reduction is only around 3% for an 85-year-old.

A model CDC scheme

To keep things simple, let’s assume that the CDC scheme has a constant investment strategy with a volatility of 10% per annum. To see the impact of the risk-sharing mechanism over all ages, we have run stochastic simulations (using our CDC simulator [1]) of a model CDC scheme with this investment strategy. We then compared the volatility of the investment returns with the volatility of members’ 'actuarial value' of accrued pension from one year to the next (grouped by age).

We have assumed that new members join the scheme each year, which together with simulated deaths provides a stable membership profile akin to what we might expect from an established whole-of-life CDC scheme.

More risk for younger members – is this a good thing?

The membership profile is a key factor in how the risk is shared. The chart shows that, although the volatility of the investment returns in the model remains constant, this investment risk is not shared evenly across all members. Younger members have significantly more investment risk than older members. Is this a good thing?

In one sense, yes. Older pensioners have shorter time horizons, which reduces their ability to ride out volatility. This risk-sharing mechanism in a CDC scheme means older members take much lower investment risk, but to balance this, younger members of the scheme take on much more risk. Investment risk-sharing is not risk-reducing overall. The results show that, by having a consistent pension increase award across the whole membership, the level of risk-sharing is very significant, with the youngest members taking on around 2.5x the risk of the current investment strategy of the scheme. Accrued benefits will start out small for younger members who only have a few years of accrual and will build up over time through their working life. This is one reason for the steepness of the investment risk-sharing profile.

Does the younger member get rewarded today for taking the additional risk through additional expected return? The answer depends on how the liabilities are discounted. If returns are as expected, the pension increase remains unchanged and the value of accrued benefits increases due to an unwinding of one year of the discount rate, which should be set equal to the expected return. If the discount curve reflects the members’ risk profiles, then younger members are being rewarded for taking more risk today, but a perfect calibration is far from straightforward. At the other extreme, if a single discount rate is used, reflecting the expected return of the overall strategy, younger members will take significantly more risk for the same expected return. The reality could sit somewhere in the middle, where some level of investment risk is shared across generations.

Deferred gratification?

So why would the younger members be willing to take on this additional investment risk without this being fully reflected in a higher expected return? One rationale is that they are taking on more risk now, in the anticipation that when they reach old age they will benefit from a much more attractive risk/return profile. There is, of course, no guarantee of this. For one thing, this requires the scheme to remain open, with regular new joiners to ensure a stable membership.

One of the benefits of investment risk-sharing is that it may, depending on the design, allow more risk to be taken overall than would otherwise be palatable, and potentially facilitate higher expected returns on assets throughout the members’ whole journey (pre- and post-retirement).

Conclusion

Younger members in a CDC scheme take significantly more investment risk than older members. This could allow the scheme to accommodate a high allocation to growth assets throughout members’ journeys (pre- and post-retirement), due to investment risk-sharing.

However, the more investment risk-sharing that is built into the design, the greater the need for confidence that the scheme will stand the test of time. That means operating in the very long term to ensure younger members today will reap the benefits from being an older member in a CDC scheme in years to come.