CGT with GIA

EdSwippet

billy2shots said:
The negative is all of that unrealised gain is also compounding up so the future CGT bill is growing.

At least that was my take on it. Without modeling it full time against my initial way I won't know lol because I can't do the math..

Let's simplify. £700k over two years at 10% growth. Assume a higher rate taxpayer, so 20% tax on capital gains, and £12k annual CGT allowance for simplicity. Also, ignore the 30-day bed and breakfast rule and assume money can be reinvested immediately (in practice you'd use two different but equivalent funds to actually do this).

Method A, pay all CGT annually. At the end of year 1 you have £770k, so £70k capital gains. Sell everything, pay CGT, and repurchase with what remains. You would pay tax at 20% on £70k less £12k allowance, £11.6k, leaves you with £758.4k invested, no unrealised gain. At the end of year 2 you have £834.24k. Pay tax at 20% on £75.84k less £12k allowance, £12.77k, leaves you with a net £821.47k.

Method B, defer CGT. At the end of year 1 you again have £770k. Sell £132k (17.14% of your shares) and repurchase. This realises exactly £12k of gains, so zero CGT, £770k still invested, and the cost of what you now hold is £580k (the 82.85% part of your original investment that you didn't sell) plus £132k (your new reinvestment), so £712k. At the end of year 2 you have £847k. Sell everything. Final capital gain is £135k, tax at 20% on that less £12k allowance is £24.6k, leaves you with a net £822.4k.

So on these assumptions you come out nearly £1k better off with method B. Over ten years, the effect of compounding will be even larger. Notice that even though your year 2 CGT with method B is larger than the sum of the two CGTs that you paid with method A, your net outcome is still better.

billy2shots

Many thanks for that. That is far simpler to understand.

One slight issue though, again probably my poor math skills -

Final capital gain is £135k, tax at 20% on that less £12k allowance is £24.6k, leaves you with a net £822.4k.


If the final gain is £135k, wouldn't that mean a CGT bill of £27k less £12k allowance = £15k bill rather than £24.6k?

Even if not, your explanation shows a £1k win over my method on only a 2 year timescale so certainly the best way to go on 5+ years.

Jeremy535897

You deduct the annual exemption from the gain, not the tax.

billy2shots

Jeremy535897 said:

You deduct the annual exemption from the gain, not the tax.

And that my friends has been my problem all along. 

I was always doing it the other way around i.e

£20,000 tax bill minus my £12k allowance= £8k final CGT bill. 


Suddenly the all is clear. 

Thanks everyone for holding my hand through this. 

EdSwippet

billy2shots said:

Many thanks for that. That is far simpler to understand.

I wouldn't call it simple exactly(!), but it at least illustrates the point.

Another way to look at it is to realise that trimming your amount invested by paying CGT earlier than necessary cuts down your future dividends. Because I've been at this for a while, I hold funds with unrealised gains approaching 200%. There's a large pending CGT bill there, but if I paid it now I'd receive around 13% less in dividends. Dividends are a large part of my income, so taking that hit just to simplify paperwork isn't sensible. I have spreadsheets that take care of all of the actual calculations.

Pay attention to the assumptions I used. Illustrative, but they may not be valid for you. For example, 10% (your initial suggested value) is perhaps too high; 7% might be closer to the long-term average return for stocks, and lower for bonds. Also, CGT is 10% in basic rate tax. If you have the basic rate tax band headroom, that could argue for paying some CGT at that rate, even if early, if it means avoiding paying 20% later on. It would take quite a while for the effect of compounding and deferral to overhaul a jump from 10% to 20% tax.

Finally, unless you're willing to wait 30 days between sale and repurchase, you'll need to use at least two equivalent investment vehicles (fund, ETF, ...). This is easy to do with broad based tracker funds or ETFs, but harder or impossible with single company shares or boutique funds. (Of course, sensible investors use only broad based trackers. ;-) )

billy2shots

All very sensible. 

The numbers I used where for ease of modeling.

My long term real figures are 6% return or even 1% above inflation.

2 global trackers will be used to avoid being out of the market for 30 days, I come version to make the income/dividends calc a bit easier. 

I am hoping that funds split over 2 GIAs his and hers and bed and ISA each year (£40k) will hopefully minimise things greatly.