Salemicus comments on Financial Effectiveness Repository - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (62)
This is not investment advice and should not be relied on as such.
You need to consider tax when you make investment. If you do not, you are likely to invest badly. (Note: this discussion uses UK tax rates and concepts. The tax system may well be different in your jurisdiction, but the same general principles will likely apply.)
For example, suppose I invest £100,000 in the stock market for 10 years. Let's say that nominal returns are 7% (5% capital, 2% dividends, which I reinvest), and inflation is 2%. Therefore naively, I might expect to end up with £197,000 after 10 years, i.e. £161,000 in 2014 money. That's a 5% real return per year, which looks like a good investment.
But now let's say I have to pay SDRT at 0.5% on share purchases, and 38% on dividend income. Then, when you sell the shares you have to pay 28% in Capital Gains Tax (with an £11k exemption). So now you only end up with £131,000, i.e. £107,000 in 2014 money. That's a 0.7% real return per year, which looks like a terrible investment.
If I had invested in a more tax-efficient way (e.g. through an ISA, and with a fund reinvesting the dividends for me so they never become income) then I would have been able to get the full 5% real return. But I can only put £15,000 per year into an ISA, so this isn't a panacea.
Note that the difference between the 5% real return and the 0.7% real return is an order of magnitude. If I had invested in something with just half the real return of the stock market, but done so in a tax-advantaged way, it would be a better investment than investing in the stock market in a naïve way. In particular, there is no Capital Gains Tax on your primary residence in the UK.
Executive summary:
And so, on a meta-level:
I get a very different result when I run these numbers. I'm not from the UK so I may be interpreting the tax rules incorrectly, but here's the logic chain I used to model it (year one, so that it can be duplicated and logic verified);
At the end of year ten, I expect 182334.10 in my account, which I then sell/crystallize.
The amount of tax I would pay is based on the "adjusted cost base" (google couldn't answer what term is used in the UK, but basically the amount paid for the shares). I work this out to be 118275.5, essentially buying the dividend-reinvested shares at the expected 5% growth rate during the 10 years. I then subtract the total stamp taxes paid, which I work out to be 580.43. This gives me a tax burden of 182334.10 - 118275.5 - 580.43 = 63478.17. I then deduct the 11,000 capital gains allowance. This results in a tax payment of 52478.17*0.28 = 14693.89.
Final net value of the sale is therefore 182334.10-14693.89 = 167640.21
In real terms, this is 134996.70, or a real rate of return of 3.2%
I believe the major difference in our numbers comes from the tax rate on gains.
In terms of sensitivity, I get a return of 4.84% with zero taxes (dividend and capital gains), whereas an increase to 9% (each div and cap rates at +1%) expected return gives 4.2%. This is under the assumption of the worst possible tax rates and scenarios (full sale at maximum rates). There are lots of timing methods to reduce the tax impact.
Granted, this is based on my knowledge of Canadian taxes, but I verified the UK equivalents as best I could and the differences are primarily in the stamp taxes. Most of the data is from https://www.gov.uk/tax-sell-shares/work-out-your-gain (and the related links).
It is very possible made a mistake in my spreadsheet. The numbers were intended for illustration only. Thanks for the correction.
You make a good point regarding timing taking gains. This is another way that thinking about tax can be very important.