Giter Site home page Giter Site logo

Comments (13)

tellme-why avatar tellme-why commented on May 23, 2024 1

Hi @mattjgalloway , thank you very much for coming back on this.

I think that:

  • DIVIDEND 1/5/2022 for 10 shares should apportion as allowed expenses to 1) BUY 3/3/2022 5 shares and 2) Leftover BUY 1/1/2022 5 shares.
  • DIVIDEND 1/5/2022 for 10 shares does not matter with the disposal SELL 1/3/2022.
  • Assume there is one more disposal SELL 1/3/2025 5 10, 50% of "DIVIDEND 1/5/2022 for 10" will become an allowable expense to that disposal.

Quote:
"And so you would assume you account for that by having allowable expenses of £5 on the disposal on 1st March.
But you'll notice there you've gone across a tax year. So if you did the tax return before you received that dividend on 1st May, you would not account for that dividend and then would do it later."
Unquote

I think this is correct; but assume you had DIVIDEND 1/5/2022 for 12 (in lieu of 10 of your example), this is tricky because DIVIDEND for 2 shares would have been allowed expenses to the SELL 1/3/2022 FOO 5 10 0, which however is kind of gone that year because you have already filed the tax return.
Option 1) In this case I would simply roll over such allowable expense to the next available disposal of FOO - no matter the number of shares of such disposal.
You lose out time value of money (especially if the next disposal is say in 2030), but It's a logical thing to do and I would think HMRC would allow it.
Option 2) Alternatively you could add such expense to any CG computation filed with the next SA108 tax return (i.e. when you have any CG - security event that could arise from any other share in your portfolio - not just FOO).

This is similar to having the disposal of all shares on 4/4/2022, but then getting a Dividend in the new tax year. Since all shares have been disposed of, the only way to recover that Dividend as the allowable cost is to do it manually in the next tax year?

Sorry if I didn't get your example - in fact I learned about B&B rule to stay clear of it, but if I haven't understood your point - please write back and I will be happy to look into it again.

FYI I previously posted into https://community.hmrc.gov.uk/ - they take a few days to respond and at times responses are not super accurate - but one can get some free help there..
Kind Regards

from cgtcalc.

mattjgalloway avatar mattjgalloway commented on May 23, 2024

Thanks for filing this! Dividends are a tricky one. I'll look into this. Does indeed look a bit wonky here!

from cgtcalc.

mattjgalloway avatar mattjgalloway commented on May 23, 2024

@tellme-why - do you have a theory on how your suggested transactions should be calculated by any chance? I'm just trying to work out even how the best way to account for this case is.

It's clearly doing something wrong right now - I just am wondering the best way to account for these transactions.

from cgtcalc.

tellme-why avatar tellme-why commented on May 23, 2024

Hi @mattjgalloway thank you for your reply and your work! :)
I am no expert at all - so please take my view with two pinches of salt.

I see one potential issue with the code:

  1. Any Dividend arising in fiscal year n, should affect the capital gains of the sales arising in current and future fiscal years (i.e. n, n+)

  2. Any Dividend arising in fiscal year n, should not affect the capital gains of sales arising in fiscal years prior (i.e n-1, n-2 etc.).

  3. When a Dividend event happens in year n, the code should lower the cost of buy (section 104 pot) according to your FIFO rule in your readme doc.
    However, this lowered cost of the new section 104 pot should only be used to calculate CG for fiscal years n and n+;
    it should not affect the section 104 pot cost of fiscal prior to n for sake of CG calculations of fiscal years prior to n.

I created a more straightforward test case to play with and spot the error:

input
BUY 18/5/2019 FOO 20 1 0
SELL 18/7/2019 FOO 10 1 0
BUY 23/5/2020 FOO 20 1 0
SELL 18/7/2020 FOO 10 1 0
BUY 18/5/2021 FOO 20 1 0
SELL 18/7/2021 FOO 10 1 0
DIVIDEND 1/6/2022 FOO 30 0 <----

output

TAX YEAR DETAILS

TAX YEAR 2019/2020

  1. SOLD 10 of FOO on 18/07/2019 for GAIN of £0
    Matches with:
  • SECTION 104: 20 at cost basis of £1
    Calculation: (10 * 1 - 0) - ( (10 * 1) ) = 0 :Note the Section 104 cost is 1

TAX YEAR 2020/2021

  1. SOLD 10 of FOO on 18/07/2020 for GAIN of £0
    Matches with:
  • SECTION 104: 30 at cost basis of £1
    Calculation: (10 * 1 - 0) - ( (10 * 1) ) = 0 :Note the Section 104 cost is 1

TAX YEAR 2021/2022

  1. SOLD 10 of FOO on 18/07/2021 for GAIN of £0
    Matches with:
  • SECTION 104: 40 at cost basis of £1
    Calculation: (10 * 1 - 0) - ( (10 * 1) ) = 0 :Note the Section 104 cost is 1

Let's now change the amount of 1/6/2022 dividend from 0 to 10:

input
BUY 18/5/2019 FOO 20 1 0
SELL 18/7/2019 FOO 10 1 0
BUY 23/5/2020 FOO 20 1 0
SELL 18/7/2020 FOO 10 1 0
BUY 18/5/2021 FOO 20 1 0
SELL 18/7/2021 FOO 10 1 0
DIVIDEND 1/6/2022 FOO 30 10 <----

Result:

TAX YEAR DETAILS

TAX YEAR 2019/2020

  1. SOLD 10 of FOO on 18/07/2019 for GAIN of £0
    Matches with:
  • SECTION 104: 20 at cost basis of £1
    Calculation: (10 * 1 - 0) - ( (10 * 1) ) = 0

TAX YEAR 2020/2021

  1. SOLD 10 of FOO on 18/07/2020 for LOSS of £2
    Matches with:
  • SECTION 104: 30 at cost basis of £1.11111
    Calculation: (10 * 1 - 0) - ( (10 * 1.11111) ) = -2 :Notice that the section 104 cost changed due to the 1/6/2022 dividend. Whislt this is correct for sake of CG arising on and after fiscal years commencing 6 April 2022, it's not correct for the sake of CG calculation arising prior fiscal years? (happy for you to challenge this assumption). It won't be correct otherwise - for example, if I file SA for 21/22 and report CG arising from see in between 6 April 21 and 5 April 22. A dividend paid on 7 April 22 should not change the CG for sells before 5 April 22 otherwise I would have to amend the SA already submitted for 21/22.

TAX YEAR 2021/2022

  1. SOLD 10 of FOO on 18/07/2021 for LOSS of £3
    Matches with:
  • SECTION 104: 40 at cost basis of £1.22222
    Calculation: (10 * 1 - 0) - ( (10 * 1.22222) ) = -3 :Notice that the section 104 cost changed due to the 1/6/2022 dividend. Whilst this is correct for the sake of the fiscal year commencing 6 April 2022, it's not correct for prior fiscal years

from cgtcalc.

tellme-why avatar tellme-why commented on May 23, 2024

sorry not sure why the site formatted my response in big/bold - anyway I hope it's readable.. :)

from cgtcalc.

mattjgalloway avatar mattjgalloway commented on May 23, 2024

Hi @tellme-why, thanks for the reply!

To be honest, this is why I largely try to avoid accumulation funds! Income class funds only attract capital return that needs accounting for, but those are much easier to work out what to do as you're just lowering the acquisition cost. The problem with how the logic works currently is that I just made dividends simply raise the acquisition cost. But you've proven that has a pretty obvious edge case :(.

I'm not sure I fully agree that it's related to fiscal year though. It's more that if the dividend happens after a sale, then that dividend cannot affect anything about the previous sales. The problem I'm having with trying to work out how to account for it, is how best calculate how to apportion dividends amongst FUTURE sales only. It seems a complete minefield tbh!

from cgtcalc.

tellme-why avatar tellme-why commented on May 23, 2024

Hi @mattjgalloway thank you very much for your reply - and apologize - I incorrectly phrased my comment "lowered the 104 section acquisition cost" whilst it should have been "increased the 104 section acquisition cost" as you suggested.

Unfortunately, the calc complexity does not go away if you only own Overseas Distributing Funds. In fact, it becomes more complex! Overseas Distributing Funds do accrue Excess Reportable Income "ERI" that is deemed (and accumulated) within the fund, i.e. is a non-cash event. This shall be treated as an "Accumulating unit Dividend Event", although it's normally much smaller than a normal distribution. With these funds, you then have to keep on check both Distributions and ERI. Distributions may classify as income or Dividends depending on fund holding<40% bonds - some online fund supermarkets may not separate them out - so you need to DIY .. :(

Hence I assume people avoid complexity and take the hit not claiming back this incurred cost. Whether people -or even accountants- understand what ERI is - and the fact you have to pay Income/Divi tax on it it's another matter.

Yes the issue may be related to dividend arising after the sale; this is more prominent for funds rather than shares as funds deemed distributions can happen 6 months after the reporting period end. One may sell just after the reporting period end - and have 6 months to go before getting to the distribution event. You end up paying income / divi tax on the distribution but have no CG to claim the cost against anymore?

from cgtcalc.

mattjgalloway avatar mattjgalloway commented on May 23, 2024

@tellme-why Oh wow ERI OK that's something new for me actually. I will read more about it!

I'll also keep thinking how best to account for all this stuff in the calculations by the way.

Back to your example, but I'll add another sell event after the dividend:

BUY 18/5/2019 FOO 20 1 0
SELL 18/7/2019 FOO 10 1 0
BUY 23/5/2020 FOO 20 1 0
SELL 18/7/2020 FOO 10 1 0
BUY 18/5/2021 FOO 20 1 0
SELL 18/7/2021 FOO 10 1 0
DIVIDEND 1/6/2022 FOO 30 10
SELL 2/6/2022 FOO 10 1 0

I am guessing this is clear that we would apportion 10/30 of the dividend of 10 as allowable expense on the final sale, right?

Then let's get sneaky and add some more sales:

BUY 18/5/2019 FOO 20 1 0
SELL 18/7/2019 FOO 10 1 0
BUY 23/5/2020 FOO 20 1 0
SELL 18/7/2020 FOO 10 1 0
BUY 18/5/2021 FOO 20 1 0
SELL 18/7/2021 FOO 10 1 0
DIVIDEND 1/6/2022 FOO 30 10
SELL 2/6/2022 FOO 10 1 0
SELL 1/7/2022 FOO 10 1 0

Again, I think clear, we apportion 10/30 to each of the last 2 sales.

What if we then add a buy that B&Bs with the final sale:

BUY 18/5/2019 FOO 20 1 0
SELL 18/7/2019 FOO 10 1 0
BUY 23/5/2020 FOO 20 1 0
SELL 18/7/2020 FOO 10 1 0
BUY 18/5/2021 FOO 20 1 0
SELL 18/7/2021 FOO 10 1 0
DIVIDEND 1/6/2022 FOO 30 10
SELL 2/6/2022 FOO 10 1 0
SELL 1/7/2022 FOO 10 1 0
BUY 2/7/2022 FOO 10 1 0

Do we then need to not apportion the dividend to this sale? Because it matches against the buy, so those shares sold are not in any way ones that were part of the dividend right?

Then I start thinking what about if you add in things like multiple dividends, selling part of a section 104 holding in between those 2 dividends. The maths just starts to completely blow my mind. I simply cannot think of how best to handle all this!!

from cgtcalc.

tellme-why avatar tellme-why commented on May 23, 2024

@mattjgalloway here is an example of ERI https://www.vanguardinvestor.co.uk/investing-explained/general-account-tax-information I quote the top left section "If you hold Vanguard Irish-Domiciled funds (including our ETFs) (also known as reporting funds) in your Vanguard general account, then you may need to declare something called "Excess reportable income". An example of fund may be a plain vanilla FTSE 250 ETF that is IRL domiciled - rather than GB domiciled - hence attracts the additional complexity of ERI. You avoid ERI if you hold only funds with ISIN commencing by "GB", but you loose variety - Vanguard do not have decent ETF for FTSE that are "GB"..

I have to think your 3 steps above are all correct!

from cgtcalc.

mattjgalloway avatar mattjgalloway commented on May 23, 2024

@tellme-why Sorry for not updating here. I have on-and-off been thinking about how to handle all this. I might need to significantly refactor how the offsetting for dividends and capital return works. But I'll do it if necessary.

I'm just wondering now about case like this:

BUY 1/1/2022 FOO 20 10 0
SELL 1/2/2022 FOO 10 10 0
DIVIDEND 1/3/2022 FOO 10 10
DIVIDEND 1/4/2022 FOO 10 10
SELL 1/5/2022 FOO 10 10 0

I guess the logic here is easy and we can just accumulate the two dividends to be a total of £20 allowable expenses on the disposal on 1st May.

Then what about not a full sale:

BUY 1/1/2022 FOO 20 10 0
SELL 1/2/2022 FOO 10 10 0
DIVIDEND 1/3/2022 FOO 10 10
DIVIDEND 1/4/2022 FOO 10 10
SELL 1/5/2022 FOO 5 10 0

I guess the logic here is again easy and we say it's half the dividends go to that sale, so £10 allowable expenses on the disposal on 1st May.

Then what about:

BUY 1/1/2022 FOO 20 10 0
SELL 1/2/2022 FOO 10 10 0
DIVIDEND 1/3/2022 FOO 10 10
DIVIDEND 1/4/2022 FOO 10 10
SELL 1/5/2022 FOO 5 10 0
DIVIDEND 1/6/2022 FOO 5 5
SELL 1/7/2022 FOO 5 10 0

The first disposal seems clear at £10 allowable expenses on the sale on 1st May again. And then second disposal has £15 allowable expenses because of the remainder from the first two dividends, and the whole of the third dividend.

Then what about:

BUY 1/1/2022 FOO 10 10 0
SELL 1/3/2022 FOO 5 10 0
DIVIDEND 2/3/2022 FOO 5 5
BUY 3/3/2022 FOO 5 10 0
SELL 1/5/2022 FOO 10 10 0

We have a B&B match there (1st March disposal, 3nd March acquisition). The B&B has a dividend in the middle. In this case I would say there's no dividend to match onto for that B&B disposal.

But here's the kicker... what about:

BUY 1/1/2022 FOO 10 10 0
SELL 1/3/2022 FOO 5 10 0
DIVIDEND 2/3/2022 FOO 5 5
BUY 3/3/2022 FOO 5 10 0
DIVIDEND 1/5/2022 FOO 10 10

Here you've got a dividend that does match with the shares bought in the acquisition on 3rd March. And so you would assume you account for that by having allowable expenses of £5 on the disposal on 1st March. But you'll notice there you've gone across a tax year. So if you did the tax return before you received that dividend on 1st May, you would not account for that dividend and then would do it later. Or even, you could have more and more dividends coming in for those shares in perpetuity which surely would have to be accounted for on that B&B'd disposal.

Options I can see:

  1. Indeed the dividends that come in after keep changing the disposal on 1st March 2022's allowable expenses.
  2. Ignore dividends across tax year boundaries and if a disposal was in a previous tax year to a dividend then tough luck, we don't apportion allowable expenses.

@tellme-why - this is the bit I'm struggling with. Do you have any thoughts here? It's the B&B that makes things confusing!

from cgtcalc.

mattjgalloway avatar mattjgalloway commented on May 23, 2024
  • DIVIDEND 1/5/2022 for 10 shares should apportion as allowed expenses to 1) BUY 3/3/2022 5 shares and 2) Leftover BUY 1/1/2022 5 shares.
  • DIVIDEND 1/5/2022 for 10 shares does not matter with the disposal SELL 1/3/2022.
  • Assume there is one more disposal SELL 1/3/2025 5 10, 50% of "DIVIDEND 1/5/2022 for 10" will become an allowable expense to that disposal.

I'm not sure I agree that the "DIVIDEND 1/5/2022 for 10 shares" doesn't matter with the disposal "SELL 1/3/2022", given it's B&B. Because the SELL 1/3/2022 is matching with the BUY 3/3/2022. And the "DIVIDEND 1/5/2022" is most definitely for those shares bought on 3/3/2022.

On the matter of a dividend affecting a previous year's tax return if things go over a tax year - the more I think about it, the more I think that actually that's the case and you should amend a tax return. I had a similar issue once where I realised that if you have an ex-div date in one tax year, but payment in next tax year (very easily done) then you probably will not get the value for equalisation payment until after your tax return is due. I posted about that here: https://www.reddit.com/r/UKPersonalFinance/comments/nt3r2k/cgt_on_fund_sale_with_equalisation_where_sale/

This really is a) fun, b) messy!

But anyway - I think I have quite a lot to go on here. One thing I'd like to do @tellme-why is to write some unit tests for these cases we're talking about. If you have any bandwidth to write out the calculations for some of these cases, I'd really appreciate it. If not I'll just request that you look over what I write for the calculations just so that we're aligned!

from cgtcalc.

tellme-why avatar tellme-why commented on May 23, 2024

Hi @mattjgalloway thank you for your reply and apologize for late response.
I have posted in the HMRC community the edge case above and looking for a response - hopefully, it may be help.
I can add one more thought without laboring it too much.
If you account DIVIDEND 1/5/2022 in the 21/22 tax return allowable expenses, the net effect is reducing your tax paid for 21/22. On the other hand, one pays tax on DIVIDEND 1/5/2022 on the 22/23 tax return, say one year later cash out event. I think that the basic principle of allowable expenses rules for CG is to avoid double taxation on Dividends; one pays dividend tax once, but not CG tax on the same amount. Therefore rolling the Dividend paid CG allowance to 22/23 seems fairer in terms of cash balance timing, namely in 22/23 one pays 1) Dividend tax on DIVIDEND 1/5/2022 and 2) Avoid double taxation by the way of CG allowance for the SELL 1/3/2022 by clawing back the CG tax paid via allowable DIVIDEND adjustment in the 22/23 tax return. The latter also saves the accountant cost of having to redo the 21/22 tax return.

Sorry I still don't get the B&B rule, I need to read it more; in this edge case made things a bit more complicated to untangle.

I'll try to pin down some more test cases..

Thanks!

from cgtcalc.

tellme-why avatar tellme-why commented on May 23, 2024

Hi @mattjgalloway

I checked with HMRC who advised:
"Hi, A deemed distribution will be treated as a taxable dividend. This is deemed to be received on the fund distribution date. You have until 31 October each year to submit a paper tax return and 31 January the following year for an online return.
Box 20, page 8 on SA100 and its online equivalent can be ticked where provisional figures have been reported.
It is possible to amend tax returns once they are submitted, either in writing to HM revenue and customs BX9 1AS, detailing the required changes or by amending the online version".
So yes there is a mechanism to make good the allowable cost paid in the next tax year...

from cgtcalc.

Related Issues (10)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.