Which platform for separate crystallised and uncrystallised pots?

TheTelltaleChart

Pat38493 said:

Further, even when accepting your POV, as soon as you start to take into account that having the equities in higher marginal tax wrappers will be statistically more likely to put you into a higher tax bracket at some future point, my point one actually becomes "yes" but not for the reason I was stating.

Sometimes, yes. Though we all tend to imagine that other people have vaguely comparable finances to ourselves, when in fact most are either a lot poorer or a lot richer, and may also have a very different split between tax wrappers. So it very much does depend.

So for some, growing uncrystallised pots could reach the TFLS limit. For others, that's not a factor, but growing crystallised pots (+ growing 75% of uncrystallised pots) could lead to a higher tax rate on taxable withdrawals. Or both. Generally, all these things do, as you say, suggest putting lower expected growth assets in the pots where the effective tax rate could rise.

It's perhaps worse than that, though. My stated assumption, "that the tax application ratios for each account is unchanged", is a stronger assumption than just saying that the marginal tax rate doesn't change. E.g. if you know your marginal rate will be 20%, but will pay a mixture of 0% and 20% (or if you know the marginal rate will be 40%, but will pay a mixture of 20% and 40%), then my assumption doesn't hold.

The problem is that, if you have a £100k crystallised pot, and expect to get £50k out at 0% tax (using your Personal Allowance), the rest at 20% tax, so the net value is £90k and the tax ratio is 0.9, then that tax ratio will change if the pot's investments grow faster than does the amount you're able to take out tax free. So if the pot grows to £125k but you can still only get £50k out tax free, then the net value is £110k and the tax ratio is down to 0.88.

At this level of complication, something like the cash flow modelling software you mention might be necessary.

Pat38493

TheTelltaleChart said:

Pat38493 said:

Further, even when accepting your POV, as soon as you start to take into account that having the equities in higher marginal tax wrappers will be statistically more likely to put you into a higher tax bracket at some future point, my point one actually becomes "yes" but not for the reason I was stating.

Sometimes, yes. Though we all tend to imagine that other people have vaguely comparable finances to ourselves, when in fact most are either a lot poorer or a lot richer, and may also have a very different split between tax wrappers. So it very much does depend.

So for some, growing uncrystallised pots could reach the TFLS limit. For others, that's not a factor, but growing crystallised pots (+ growing 75% of uncrystallised pots) could lead to a higher tax rate on taxable withdrawals. Or both. Generally, all these things do, as you say, suggest putting lower expected growth assets in the pots where the effective tax rate could rise.

It's perhaps worse than that, though. My stated assumption, "that the tax application ratios for each account is unchanged", is a stronger assumption than just saying that the marginal tax rate doesn't change. E.g. if you know your marginal rate will be 20%, but will pay a mixture of 0% and 20% (or if you know the marginal rate will be 40%, but will pay a mixture of 20% and 40%), then my assumption doesn't hold.

The problem is that, if you have a £100k crystallised pot, and expect to get £50k out at 0% tax (using your Personal Allowance), the rest at 20% tax, so the net value is £90k and the tax ratio is 0.9, then that tax ratio will change if the pot's investments grow faster than does the amount you're able to take out tax free. So if the pot grows to £125k but you can still only get £50k out tax free, then the net value is £110k and the tax ratio is down to 0.88.

At this level of complication, something like the cash flow modelling software you mention might be necessary.

OK well now if I take a step back, I could make the case that it's a waste of effort to make such small adjustments as, for example, £1515.

Let's say I have a £100K pot which is £80K in drawdown and £20K in ISAs.  My chosen target risk tolerance / allocation is 80%20% Equities / Cash.

I took the easy way out and just put all the drawdown in equities and all the ISA in cash.  My allocation is 80/20.  Wrong - it is actually 76.1% / 23.8% per your calculations and within the assumptions described.

However, my error in equity allocation is less than 4%.  

My new lazy theory is - unless I believe that my risk tolerance allocation choices are accurate to within 4% or less, there is no point making such adjustments.

Most investors are encouraged to choose an allocation of 80/20, 70/30, 60/40 or whatever, so they are effectively choosing the equity allocation with an accuracy of +/- 10%.  As such, and error of 3.8% is irrelevant - I am polishing a turd to use the colloquial expression?

Further - if I look at detailed studies on safe withdrawal rates like, for example, the ERN blogs, they generally tested rates at increments of 10%.

Therefore, it's entirely plausible that my "wrong" allocation of 76.1% is actually "better" than your exactly adjusted net 80/20 allocation.

Also taking into account all your points above, I think it would be too complicated for me to bother with such adjustments, since my allocations are currently estimated using time windows into the future like:
2 years cash
3-5 years 70/30
6-10 years 80/20
10+ years 100% equities.

If I then try to calculate the adjustments across all my pots all the time to equalise the net values, the adjustments will be too small to be outside the general opening margin of error.

You are also right that for me personally I am in danger of paying 40% tax if my pension pots grows too much, so I am generally planning to bring money out of the pensions within the 20% tax bands, at least for the first 10 years or so.

Cobbler_tone

TheTelltaleChart said:

You merge 2 pots, one with gross assets P1 and tax application ratio t1, the other with P2 and t2.

The tax application ratio for the merged pot can be calculated as:

t3 = ( (t1 * P1) + (t2 * P2) ) / (P1 + P2)

Now consider an asset class held in either or both pots, with gross values of g1 and g2 in the 2 pots respectively.

Before the merger, the net value in this asset class is:

(g1 * t1) + (g2 * t2)

Assuming that's the net value you want, you now need a gross value of:

( (g1 * t1) + (g2 * t2) ) / t3

So you need to buy another:

( (g1 * t1) + (g2 * t2) ) / t3 - (g1 + g2)

of it inside the merged pot (and if that's a negative figure, it indicates a sale).

In the specialised case when this asset class only features in the first pot, not in the second one (i.e. when g2 = 0), the above can be simplified, as follows …

After the merger, you need a gross value in this asset class of:

g1 * t1 / t3

So you need to buy another:

g1 * t1 / t3 - g1

or

g1 * ( t1/t3 - 1)

of it.

OK Stephen Hawkins.
How did we end up from "HL have two accounts, one for each" to this? 

I think it is a good example of why most people wouldn't bother. 

Triumph13

I entirely agree that virtually no-one would actually make the £1,515 adjustments, for precisely the reasons you give. Cobbler_tone.  I only did it in the example to show how the maths works, but it would be a case of 'measure with a micrometer, cut with an axe, or whatever the opposite of that is.  A bit like when a news story takes someone's estimate of a round number of yards and converts it to metres to two decimal places.

LHW99

IvanOpinion said:

Have we figured out how much a box of cornflakes is yet?

No but it's probably related to the size of the black hole in the middle of the Milky Way by some means

Triumph13

TheTelltaleChart said:

Though we all tend to imagine that other people have vaguely comparable finances grasp of algebra to ourselves, when in fact most, are either a lot poorer or a lot richer  definitely do not.

 

Fixed that for you 

TheTelltaleChart

Pat38493 said:

My new lazy theory is - unless I believe that my risk tolerance allocation choices are accurate to within 4% or less, there is no point making such adjustments.

Yes, it's probably not worth making portfolio adjustments just for this. But the next time you're rebalancing, or making other significant changes, you could take into account the changes as a result of merging pots. By which, I don't mean look backwards at the effect of the merger, but use the net values of all holdings when reckoning their current contribution to the portfolio.

E.g. I have an uncrystallised pot, on the taxable 75% of which I might end up paying a mixture 0% or 20%. So the net value is somewhere betwen 0.85 and 1 of the gross value. I've been using 0.925 as a middle-of-the-road figure. So when I'm checking/rebalancing my whole portfolio (including ISA and pension), I'm treat each £1k inside the pension as being worth £925.

One time, I forgot to multiply by 0.925, and so I bought the "wrong" amount of a holding! So I just left it as it was 🤷

Pat38493

TheTelltaleChart said:

Pat38493 said:

My new lazy theory is - unless I believe that my risk tolerance allocation choices are accurate to within 4% or less, there is no point making such adjustments.

Yes, it's probably not worth making portfolio adjustments just for this. But the next time you're rebalancing, or making other significant changes, you could take into account the changes as a result of merging pots. By which, I don't mean look backwards at the effect of the merger, but use the net values of all holdings when reckoning their current contribution to the portfolio.

E.g. I have an uncrystallised pot, on the taxable 75% of which I might end up paying a mixture 0% or 20%. So the net value is somewhere betwen 0.85 and 1 of the gross value. I've been using 0.925 as a middle-of-the-road figure. So when I'm checking/rebalancing my whole portfolio (including ISA and pension), I'm treat each £1k inside the pension as being worth £925.

One time, I forgot to multiply by 0.925, and so I bought the "wrong" amount of a holding! So I just left it as it was 🤷

The point is more around what you are rebalancing to - how did you decide or do you decide your asset allocations targets.  Mostly this revolves around discussions about risk tolerance with then setting allocation in +/- 10% increments.  

As far as I can tell, even if you ignore the difference between ISA, Drawdown, UC, and just set allocation target at the gross level, you won’t get an error greater than 4-5%.  As such, the difference is within the margin of error on how you determine your original targets.

I attend Q&A sessions with an IFA who provides a kind of DIY financial planning course and then it includes a Q&A every month.  I found a few questions in some of his Q&A around this topic of rebalancing and tax rates on different assets (the Q&A sessions are really well documented so you can find relevant questions and then watch the relevant section).  Based on this I am pretty sure they would say that we are making it way too complicated and we should just pick a number and stick with it, regardless whether it’s gross, net or whatever - the important thing is not whether you have 80% or 76.2323r234% equities - the important thing is more that you stick to the plan of rebalancing based on the current situation on regular intervals.

michaels

TheTelltaleChart said:

Pat38493 said:

My new lazy theory is - unless I believe that my risk tolerance allocation choices are accurate to within 4% or less, there is no point making such adjustments.

Yes, it's probably not worth making portfolio adjustments just for this. But the next time you're rebalancing, or making other significant changes, you could take into account the changes as a result of merging pots. By which, I don't mean look backwards at the effect of the merger, but use the net values of all holdings when reckoning their current contribution to the portfolio.

E.g. I have an uncrystallised pot, on the taxable 75% of which I might end up paying a mixture 0% or 20%. So the net value is somewhere betwen 0.85 and 1 of the gross value. I've been using 0.925 as a middle-of-the-road figure. So when I'm checking/rebalancing my whole portfolio (including ISA and pension), I'm treat each £1k inside the pension as being worth £925.

One time, I forgot to multiply by 0.925, and so I bought the "wrong" amount of a holding! So I just left it as it was 🤷

Hmm. Should we also be adjusting pot valuations for (potential) inheritance tax?!

Pat38493

michaels said:

TheTelltaleChart said:

Pat38493 said:

My new lazy theory is - unless I believe that my risk tolerance allocation choices are accurate to within 4% or less, there is no point making such adjustments.

Yes, it's probably not worth making portfolio adjustments just for this. But the next time you're rebalancing, or making other significant changes, you could take into account the changes as a result of merging pots. By which, I don't mean look backwards at the effect of the merger, but use the net values of all holdings when reckoning their current contribution to the portfolio.

E.g. I have an uncrystallised pot, on the taxable 75% of which I might end up paying a mixture 0% or 20%. So the net value is somewhere betwen 0.85 and 1 of the gross value. I've been using 0.925 as a middle-of-the-road figure. So when I'm checking/rebalancing my whole portfolio (including ISA and pension), I'm treat each £1k inside the pension as being worth £925.

One time, I forgot to multiply by 0.925, and so I bought the "wrong" amount of a holding! So I just left it as it was 🤷

Hmm. Should we also be adjusting pot valuations for (potential) inheritance tax?!

This opens another can of worms I guess.  If you expect to have a large amount left over at the end of life, and you prefer to leave it to others rather than increase current spending or give it away in your lifetime, traditionally you would have left it in the pension, but this will all change next year I guess - hence why a lot of people now are re-hashing their plans to pass wealth to others before they die.  

Possibly this is a deliberate or hoped result of upcoming changes, although this would imply a lot of joined up thinking and research.

Even now, if you die before 75 the pensions will get favourable tax treatment.

Personally I am planning to draw a lot out of my pension within the 20% tax bracket for 10-12 years, partly to avoid the risk of having more money than I expected, but having it locked up and having to pay 40% tax to get it out - a good problem to have I guess but not ideal.   Also - this enables me to give away more money earlier - I am already planning to gift money for LISA contributions to our kids in the coming years.  

In other words, based on the upcoming pension IHT changes, rather than adjusting pot valuations, it’s more like you should possibly consider trying to give excess money away before you hit Davey Jones locker.