CDC is a new type of pension arrangement for the UK. In a whole-of-life CDC scheme, employers and employees pay fixed contributions (akin to traditional defined contribution, or DC), but instead of individual DC pots, members accrue target pension benefits payable from the scheme, from retirement age until death.

CDC provides an income in retirement for its members, removing a lot of the complex decision-making required in traditional DC. However, with that comes less flexibility. In this article, we explore whether CDC can provide a better outcome for members than current DC offerings.

Comparison with guaranteed annuities

Firstly, as CDC provides an income for life in retirement, let’s compare this to DC members purchasing an annuity at retirement. The CDC pension could be expected to be higher than that purchased through a guaranteed annuity for a few reasons:

Although the points above suggest higher benefit levels could be expected in CDC vs annuity purchase, the two benefits are not equivalent. The most important difference being, in a CDC scheme, the member is still exposed to investment risk post-retirement, whereas with annuity purchase, that risk is passed to the insurer. However, the risk-sharing mechanism within CDC means older members take less investment risk than younger members.

Using abrdn’s CDC simulator[1], we can stochastically simulate the running of a CDC scheme and compare the outcomes for each member with what they could have received in a traditional DC scheme with annuity purchase at retirement.

In each year of the simulation, we examine all the members that reach retirement age and compare the CDC pension they have accrued in the scheme with an estimate of the level of pension they could have received if they had instead had their contributions paid into an individual DC pot, used to purchase an annuity at retirement. The chart below shows the results for one simulation.

The x-axis is the year of the simulation (here we have run the CDC scheme over 150 years), and the y-axis shows the percentage uplift in CDC pension vs traditional DC with annuity purchase. For example, if the uplift is 25%, in our model, the member received a 25% higher starting pension at retirement in the CDC scheme than they would have received in a traditional DC scheme with annuity purchase. The annuity value used for the comparison is an estimate based on a guaranteed pension increase equal to the level of pension increases in the CDC scheme at the point of retirement.

The average uplift across all members who retired during the simulation is 22%, assuming the DC pot is invested in line with the CDC investment strategy, and this increases to 29% if we assume a typical lifestyling strategy is used for the DC pot (de-risking gradually from 10 years before retirement). The dotted lines in the chart show the average uplift for those retiring in that year of the simulation. What we can see is that the average uplift factor changes over time, dependent on the year of retirement. Those retiring in year 40 have a much better relative outcome from CDC vs DC, than members retiring in year 70, with some members appearing to be better off under traditional DC. So why is this?

The main variable impacting the results is the path of investment returns. The charts below show the 10-year moving average investment return in our 150-year simulation. If there have been particularly poor investment returns in the 10 years before retirement, the CDC uplift is higher. This is because members in traditional DC are more directly impacted by falls in investment return in the run-up to retirement. The impact is less with a lifestyling strategy, but the trend still applies. The converse is also true: strong investment performance in the 10 years before retirement results in lower, even negative uplifts for CDC vs DC.  

Therefore, although on average there is a material uplift in our simulation, the path of returns is a very significant factor. The conclusion is that CDC reduces the members’ exposure to investment returns in the run-up to retirement. 

DC drawdown

The other comparison we can make is with DC drawdown. For the members in our simulation, we have compared the total benefits paid under CDC with a DC drawdown solution. For each member that died during the simulation, at the point of death we have calculated the total pension received from CDC (from retirement until death).

We have then calculated the total pension that would have been received if the member had instead invested contributions in an individual DC pot and taken an income equal to the CDC pension payments.

If the DC pot is exhausted, the pension payments stop. At the point of death, any remaining DC pot is paid to the member (inheritance). The chart below shows the total income from CDC as a proportion of that from DC drawdown. A result of 100% implies that the same overall retirement income would have been received from both solutions. A result greater than 100% implies a higher overall income was received from CDC than DC drawdown.

The chart below shows the results. We have shown the results based on two DC investment strategies, one where the investment strategy is consistent with that of the CDC scheme, and the other where we have assumed a typical lifestyling strategy (de-risking as retirement approaches and in retirement).

In the simulation there are a group of members that die before retirement and do not receive any CDC pension. For the purpose of the model, we have assumed no death benefits are payable, and hence they would have been better off in a traditional DC arrangement where the DC pot is paid on death. In practice, we would expect some form of death benefit payable from the CDC scheme.

The chart shows a median result of around 100% when we assume the DC pot is invested consistently with the CDC strategy. This is to be expected, although as you can see, there will be winners and losers due to the path of investment returns and longevity experience. However, under the DC lifestyling assumption the median result is 150%, which is where we can see the impact on expected benefits levels from de-risking the investment strategy. Again, there are winners and losers.

Conclusion

One conclusion from both the guaranteed annuity comparison and the DC drawdown comparison is that median outcomes can be higher under CDC, primarily due to the higher growth allocation within the investment strategy. Note that our CDC design, for the purpose of the modelling, assumes investment risk-sharing across generations, allowing high growth allocations where there is a stable membership. However, CDC does not make everyone better off. The path of investment returns and longevity experience will always be a very important factor in individual outcomes. 

CDC is a very different beast to the current DC offerings, particularly DC drawdown. The differences are significant and also nuanced, with very different potential outcomes. It’s not possible to say one is better than the other. Pooling investment and longevity risk has its advantages but the winners and losers cannot be predicted in advance.

Will higher median outcomes help attract employers and members to CDC once it becomes possible? Careful communication will be important to ensure understanding of the differences between CDC and DC, and the risks associated with each.