Who can claim £5,000 tax-free interest?

polymaff

Using "old thinking" the Personal Allowances would be allocated in the same order as for the assessment "Order of Taxation". It is amazing how many HMRC staffers STILL believe this to be so.

So you'd "shade" the first £11,850 of Non-Savings Taxable Income and say "job done". That would result in the remaining £24,150 of the Non-Savings Taxable Income, all but £500 of the Taxable Savings and all but £2,000 of the Dividend Income being taxed. Total Tax bill £9,330.

Here's a suggestion which may make my model easier to use. When you've finished shading and are identifying the tax rates for the various unshaded sections of the column, write the rate in the unshaded section. Remember that if the "bar" splits an unshaded section there is a different rate to write in either side of the bar.

How you then allocate Personal Allowances in "new thinking" mode should stand out.

short_butt_sweet

polymaff wrote: »

Using "old thinking" the Personal Allowances would be allocated in the same order as for the assessment "Order of Taxation". It is amazing how many HMRC staffers STILL believe this to be so.

So you'd "shade" the first £11,850 of Non-Savings Taxable Income and say "job done". That would result in the remaining £24,150 of the Non-Savings Taxable Income, all but £500 of the Taxable Savings and all but £2,000 of the Dividend Income being taxed. Total Tax bill £9,330.

Here's a suggestion which may make my model easier to use. When you've finished shading and are identifying the tax rates for the various unshaded sections of the column, write the rate in the unshaded section. Remember that if the "bar" splits an unshaded section there is a different rate to write in either side of the bar.

How you then allocate Personal Allowances in "new thinking" mode should stand out.

ok, i guess the problem is that i'm not convinced that your visual model (in post #35) works for allocating the PA. (or that any visual model works for the PA.)

your model has a "start of higher rate" marker at a height of BRB-plus-PA. and you want to allocate the PA parallel with either that BRB-plus-PA band (below the marker) or the higher rate band (above the marker).

but AIUI, the PA isn't allocated parallel with BRB-plus-PA or HRB. (by contrast, the SRA, PSA and DA are allocated parallel with the BRB or HRB. so handling them in a visual model can work OK.)

the way i thought the PA should be handled is to take out an amount of income equal to the PA from the column altogether, with a free choice about what kind(s) of income you take out. and all income not taken out then drops down to fill the space. and the marker you have at BRB-plus-PA level should instead be at BRB level.

if we do it your way, then you are effectively widening the BRB. because somebody (with a suitable mix of income types) can have up to BRB+PA taxed at basic rates, in addition to having PA of income untaxed.

so i think your model is wrong. or am i just behind the times?

Dazed_and_confused

I've got to confess I'm struggling to follow this simple visual model

Using the new BRB of £34,500 i have no problem getting to the £9,330 as the traditional, but now wrong, total liability.

I have managed to reduce this by £106.25 to £9,223.75 but as this isn't a round hundred then I know it's not the answer polymaff believes is the lowest possible outcome.

Back to the abacus!

Dazed_and_confused

The original £9,330 consisted of income (after Personal Allowance) taxed at six different rates/band combinations* and I've got this down to just four now but cannot get the liability any lower than £9223.75.

Must try harder

*
Basic rate
PSA nil rate (within the basic rate band)
Savings basic rate
Dividend nil rate (within the basic rate band)
Dividend nil rate (within the higher rate band)
Dividends at the dividends higher rate

polymaff

short_butt_sweet wrote: »

ok, i guess the problem is that i'm not convinced that your visual model (in post #35) works for allocating the PA. (or that any visual model works for the PA.)

your model has a "start of higher rate" marker at a height of BRB-plus-PA. and you want to allocate the PA parallel with either that BRB-plus-PA band (below the marker) or the higher rate band (above the marker).

but AIUI, the PA isn't allocated parallel with BRB-plus-PA or HRB. (by contrast, the SRA, PSA and DA are allocated parallel with the BRB or HRB. so handling them in a visual model can work OK.)

the way i thought the PA should be handled is to take out an amount of income equal to the PA from the column altogether, with a free choice about what kind(s) of income you take out. and all income not taken out then drops down to fill the space. and the marker you have at BRB-plus-PA level should instead be at BRB level.

if we do it your way, then you are effectively widening the BRB. because somebody (with a suitable mix of income types) can have up to BRB+PA taxed at basic rates, in addition to having PA of income untaxed.

so i think your model is wrong. or am i just behind the times?

I'm losing it. Big time, as just before putting up that "almost" beta model I emphasised that the "new" was about the "parameter" BRB - which wasn't driving the conversation - and which I then let become a dependent variable.:(

I still think that the importance of the BRB and of the change in the part that PA plays is not generally understood and that an almost visual model of a sub-100k, three-stream is possible - but probably not worth inflicting on the world.

Useful feedback, though, from those who've shown an interest - many thanks. I'll take down the original.

My non-simple, non-visual software - only about 800 lines of code - suggests to me that the £36k, £10k, £10k analysis should be £9,223.75, so full marks, D+C

Dazed_and_confused

Thank goodness for that! Had to have a lie down in a dark room for a bit this afternoon after trying to get to a lower figure.

short_butt_sweet

polymaff wrote: »

My non-simple, non-visual software - only about 800 lines of code - suggests to me that the £36k, £10k, £10k analysis should be £9,223.75, so full marks, D+C

aha ... i've finally figured out how you get to £9,223.75.

this is news to me, in that i thought there were two general reasons why the PA is sometimes best applied in a non-traditional way (where "non-traditional" means: not following the "order of taxation"), and this example introduces a third reason.

actually, this example is a hybrid of two reasons.

the reason i already knew about is when the the traditional way leaves the DA (wholly or partly) parallel with the BRB, and by using the PA against some dividends, we can move the DA so that it's parallel with the HRB. this saves 32.5% tax on dividends, and incurs 20% tax on non-savings or savings income, a net saving of 12.5%.

so in this example, i wanted to put £350 of the PA against dividends, saving £43.75 tax.

the new reason is where the PSA would (following the traditional way) be parallel with the BRB, but by using some of the PA against savings instead of non-savings income, the PSA can be moved parallel with the HRB. in situations where the traditional way leaves some interest taxed at higher rate, this would save 20% tax on up to £500.

in this example, it's more complicated. the traditional way doesn't tax any interest at 40%, so there is no 20% saving. but we can get a 12.5% saving on an extra £500 by setting the PA against all but £500 of the interest (i.e. against £9,500 of interest), and also setting the PA against an additional £500 of dividends (so in total, it's set against £850 of dividends).

so the PA (£11,850) is now set against:
- £1,500 of non-savings income
- £9,500 of savings income
- £850 of dividends

the remaining income (not covered by the PA) is therefore:
- £34,500 of non-savings income
- £500 of savings income
- £9,150 of dividends

that non-savings income exactly fits in the BRB, and is taxed at 20%.

in the HRB, the lowest layer is £500 of savings income, which is also parallel with the PSA, so taxed at 0%.

next is £2,000 of dividends, parallel with the DA, so taxed at 0%.

finally, there is £7,150 of dividends, taxed at 32.5%.

once you've allocated the PA optimally, you can describe the BRB, HRB, SRA, PSA & DA visually. i'm struggling to imagine a visual way of allocating the PA.

polymaff

Just for the record, the output from my all-singing, all dancing but crude GUI software is given below:

bt = best. pa=PA n=nonsavings, s=savings, d=dividend, l=certain lumpsums, i,=certain life policy gains 0=PA and nil rate, 1=basic rate, 2=higher rate, sa=starting rate for savings, psa=PSA, da=dividend allowance


pnds,pnc_btpan_btpas_btpad_btpal_btpai
09223,75,01500,09500,00850,00000,00000

nonsav__savs__divis___Lump___Life
036000,010000,010000,000000,000000

__n0____n1____n2____sa____psa___s0
01500,34500,00000,00000,00500,09500

__s1____s2____da____d0____d1____d2_reliefs
00000,00000,02000,00850,00000,07150,0000