The authors – Iain Clacher and Con Keating
As a prelude to considering notional contribution allocation within a CDC scheme, it is worth asking the question as to how a 64-year-old and 24-year-old might have fared had they been in individual DC rather than in collective DC pensions arrangements.
We use the same basic pedagogic set-up as previously. ( Pension Playpen – Connecting People ) This is a scheme which has just two members, a 24-year-old and a 64-year-old. The younger earns £30,000 and the older £45,000. Retirement is at age 65 for both, and wage and price inflation are 2% pa. Both members earn a notional 1/80th of final salary for life as a retirement benefit. The contribution rate is 15% of pensionable pay.
The CDC arrangement (brown line – Diagram 1) requires an investment return of 5.67% on the total contributions of £11,250 to pay all pensions on time and in full. The contribution for the older member is £6,250 and for the younger £4,500 – these are the individual DC credited amounts. The outcomes for this are shown in diagram 1 below. We assume that the individual members can earn that same return as that needed for the CDC arrangement. This is unlikely to prove possible in practice since they lack the economies of scale and scope available to a collective fund and the scale that such funds can achieve, but it is an interesting thought experiment.
For the younger member we show also a second case, where the younger member ‘life-styles’ their fund (Yellow line), achieving this same 5.67% rate until a decade before retirement at which point the individual begins to de-risk uniformly to an all gilt, 1% yield, portfolio at and beyond retirement. This strategy is similar in many regards to the de-risked portfolios being advocated by the Pension Regulator for maturing DB schemes.
Diagram 1: Individual DC compared to a CDC scheme
It is immediately obvious that the older member (black line) runs out of funds after just two thirds of their pension has been paid. It falls far short of the 9.13% return that it needs to be able to meet the pension in full.
We can see the often-expressed case for individual DC in that the younger member following the same investment strategy, achieving a 5.67% return, as the CDC collective would be better off alone. (24-IDC, Blue line above) We have truncated the trajectory of this strategy at £50,000 – it continues above exponentially. It generates a surplus of £131,502 by the time of death of the younger member and the payment of pensions is barely discernible a deviation in its trajectory. The 5.67% return is well in excess of the required 3.98% rate.
However, if the younger individual follows the de-risking strategy described earlier, moving to gilts and a 1% return (Yellow line), they will run out of funds some two years before death. The sensitivity of the younger member’s outcomes to post-retirement investment returns may be judged from the fact that the younger member would have to have achieved a return of 5.53% pa to have sufficient funds in cash (and earning nothing) at retirement to pay the pension in full.
Of course, there are many caveats needed for this analysis. Would the individual have been able to achieve these returns? What would they have had to reserve for longevity uncertainty, and so on? There is nothing about individual DC which is collective, nor does it produce any risk-sharing.
Incidentally, this analysis illustrates one of the principal advantages and attractions of collective schemes, the absence of a de-risking imperative for open ongoing schemes, from which higher total investment returns may be obtained.
The analysis so far has been on a scheme which has a breakeven rate of 2.79% pa. The breakeven rate is the internal rate of return below which, it is the older members who support younger. It occurs naturally if contribution rates are 28.6%. For contribution rates below this (such as the 15% of the analysis above) younger members support the older, for contribution rates above, it is the older who support the younger members. This is a form of mutual insurance among the members of the scheme.
The insurance Aspect
As higher rates of wage and price inflation apply to the pension awards, the breakeven point increases. If we increase the earlier assumption of 2% wage inflation to 5% and 4% price inflation, the breakeven yield rises to 5.73% from the earlier 2.79% and the contribution rate required for breakeven allocation declines from 28.6% of salaries to 22.8%.
This means that at contribution rates above 22.8%, it is the older member who supports the younger. For example, at a 30% scheme contribution rate, the CAR or required rate of return of the scheme would be 4.80% and the required rate of return of the 24-year-old would be 5.20% while the older member has a requirement of just 3.30% ( with the allocation at 30% of salary to each member).
This is extremely valuable insurance to younger member. In good times when returns are high, and contribution rates low, the younger member supports the older. While in bad times, when investment returns are low, it is the older who supports the younger. The younger member of course has far higher dependence on investment returns than the older.
An Operational Advantage of the DB Allocation Method
The DB Approach.
The total contribution for the two members, at 15%, is £11,250. The implicit required rate of return or contractual accrual rate on this award is 5.67% pa. If we then wish to notionally allocate the total award such that both members will achieve their promised benefits, we must set the implicit rate of return for each to be equal to the scheme required rate, 5.67%. The solution is to allocate £9,375 to the 64-year-old and £1,875 to the 24-year-old. These are of course the most extreme allocations possible within the bounds of the scheme set-up. They are respectively, 20.83% and 6.26% of the members’ salaries. The scheme and member trajectories are illustrated in Diagram 2.
Diagram 2: Contribution Allocation – DB style – younger members supporting older members
The trajectories shown are the product of the initial contribution which is compounded at 5.67% from which pensions paid are deducted at the time they occur. From this diagram of the trajectories, we see that everywhere the sum of the beneficial interests of the scheme are equal to the overall scheme beneficial interest or required funding level. This means that either member may withdraw and take a cash equivalent transfer value of their beneficial interest at any point in time with no detriment to the scheme or other member(s). The scheme is inter-temporarily fair to all.
This and all of the diagrams following show the timeline (x-axis) beginning at 1, this is a case of Microsoft Excel being ‘helpful’, 1 should read 0 and all other points be one lower.
Next, we will set out the case where the older member is supporting the younger member.
Diagram 3 Contribution Allocation – DB style – older members supporting younger members
The required rate of return on assets is 3.63%, which would require a contribution of 45% of pensionable pay for the scheme. Diagram 3 shows the evolution of the scheme (red dashed line). It also shows the evolution of older and younger member’s interests under both the standard DB allocation and individual DC.
Looking first to the older member we see that investment of 45% of the member’s salary will be more than sufficient to pay the pension (dark blue line). Reducing the notional allocation from £20,250 to £12,985 shows that the elder member is provided for (light blue line). This amount is instead applied to the younger member, whose notional allocation rises from £13,500 to £20,765. This is movement from the insufficient dark green line to the light green and the scheme’s dashed red.
Obviously, the amount by which the younger member’s allocation rises is the same as that of the older member’s decline. However, though the time zero values in this diagram appear similar, they are in fact different – 64-year-old £20,250 and £12,985, 24-year-old £13,500 and £20,765.
The Individual DC (IDC) Approach
Diagram 4: Contribution allocation – IDC approach.
Here, each member is allocated 15% of their salary as the contribution. This means that the 24-year-old’s required rate of return on their £4,500 initial contribution is 3.98% pa while the 64-year-old has a required rate of return (contractual accrual rate) of 9.13% on their allocated contribution of £6,750. The 64-year-old’s trajectory (orange line) is based upon 9.13% accrual, the 24-year-old’s (grey line) on 3.98% and the scheme overall (blue line) on 5.67%.
It is immediately evident that there is a difficulty, arising from the different interest or accrual rates of scheme and members – from the 19th year onwards, the beneficial interest of the younger 24-year-old member is greater than the overall scheme interest. If the scheme were funded to this scheme level it would be reporting deficits on that basis alone. In fact, the sum of the two beneficial interests is, other than at start and end points, everywhere greater than the scheme – this is illustrated below as diagram 4 (spurious deficits). This disparity results from the differing initial allocations and the mathematical property whereby internal rates of return do not average simply. The disparity is largest proportionally in year 19, at 166%, and in amount immediately prior (year 41) to the payment of the first of the 24-year-old’s annual pensions, at £4,490. The scheme would be reporting deficits relative to the original terms everywhere.
Transfers out, which rightly take place at the net asset value of member interests, would take place at deficits to the apparently ‘promised’ benefits and would carry consequences for remaining members and the scheme. Members have an apparent incentive to transfer out unless the adjustments to their beneficial interests are correctly calculated.
Diagram 5: Spurious Deficits
It is possible to create spurious surpluses, a counterpart to this diagram, when contribution rates are high and required returns low under this individual DC type arrangement, and with that potential demands for higher distributions to pensioners. In more general terms, this illustrates the fact that a scheme may be viable in the long-term but apparently, and indeed possibly, failing in the immediate term.
