Bond Surrender Calculations
HR246
Member
Hi everyone,
I have a query regarding bond surrender calculations. I've been mulling over it for a while and found some useful bits online that seem to imply an answer, but haven't found one stated outright anywhere. I'm hoping someone can help.
Say a client needs to make a withdrawal from a bond, and it's been decided a mixture of segment surrender and partial withdrawals to utilise the 5% allowance is the way forward. I make a partial withdrawal from the bond utilising the 5% allowance. I then surrender some segments. Will the 5% allowance need to be recalculated and reduced to take account of the segments that have been surrendered (as the gain on partial withdrawals is not calculated until the end of the policy year), even if the partial withdrawal is executed before the segment surrender?
Thanks in advance!
Comments
Hi
I find the easiest way to deal with bond gain calculations is at a segment level. Each segment has its own 5% allowance and its own history of premiums and withdrawals.
The maximum 5% allowance on the entire bond is therefore the initial investment into each segment multiplied by the remaining segments. If you surrender some segments, the remaining 5% allowance will immediately reduce.
If you surrender some segments after taking a partial withdrawal from those segments, the amount of the previous withdrawals from them is added to their surrender value.
Pru have a good worked example:- https://mandg.com/pru/adviser/en-gb/insights-events/insights-library/bond-segment-surrender
Also find Quilter Chargeable gains calculator helpful:- https://platform.quilter.com/support-and-help/tools/chargeable-gains-calculator/
Steven
Thanks both. I have been trying to wrap my head around it and think I have made some progress. The Pru example is helpful, as is the Quilter chargeable gains calculator.
I have looked at an example here https://www.moneymarketing.co.uk/view-from/planning-an-exit-5-allowance-vs-total-segment-surrender/ which I have also found helpful, but it has made me wonder how, in the Case Study section, they arrived at surrendering 40 policies fully and taking partial withdrawals across the remaining 60. Why 40 and 60? Is it just trial and error and that's where they ended up? How would you approach a similar calculation in real life to ensure you've got the most advantageous (in terms of minimising gain) mix of full surrender and part surrender?
Pru also have a gain calculator that also allows you to model different outcomes based on the withdrawal amount.
https://www.mandg.com/pru/adviser/en-gb/tools-calculators/bond-gain-tool
Identifying the minimum gain isn't necessarily correct thing to do. If a larger gain through a partial has no tax impact on the individual then maintaining segments to keep the full 5% allowance, and the fact the gain gets deducted from the final surrender value could deliver a better outcome.
This needs balanced by looking at whether the withdrawals you are deferring are likely to be tax free on ultimate surrender.
Thanks Les. The Pru calculator looks really helpful and I'll definitely give it a go next time I have a bond surrender case.
I see what you mean in that identifying the minimum gain isn't necessarily the best thing to do. If that was my objective, though, is there a way you would approach a 'combination' surrender (by that I mean a mixture of partial and segment surrender) to get the smallest gain? I appreciate this may not be done often in practice as there are other considerations, but I'm just trying to get my head around how to approach these combination surrenders as they continue to mystify me slightly!
If memory serves me, the Pru calculator can do this: it's a really good piece of kt.
Hi
Only thing I can think of is ...
Go to our bond gain tool.
Start with all the intended withdrawal as a withdrawal across all segments - note the gain
Add 1 segment surrender - note the gain
Add 2 segment surrenders - note the gain
Repeat until you have surrendered the maximum segments you need to to meet the withdrawal.
Find smallest gain.
Cross fingers you only have 20 segments and not 1000 :-)