CSH2: taxation and performance

masonic

Investors in the ETF had no entitlement to the returns from the collateral basket of securities - that was sold to the swap counterparty. Investors get a share of the proceeds of the swap, whose underlying subject matter is money placed at interest or interest rates. It may be that Amundi considered the derivative was of a different nature than HMRC considered it to be, or the nature of the derivative has changed compared with prior years. At a guess, the derivative would be classed as a contract for difference relating to an index or factor, making the subject matter overnight interest rates. This is where it would be necessary to consider whether that constitutes assets in an interest-bearing (or economically similar) form. I could see that being a matter over which different opinions could be formed - a return pegged to an interest rate is not itself interest - but I am only speculating.

We don't know the details of the swap, so we don't know whether or not it includes dividends from the basket of securities, so it may be that those can make up some of the returns paid to investors. Perhaps this would be covered in the annual report. As I hold exclusively within tax shelters, I am not sufficiently motivated to go looking. However, if this was the case, that could be an argument for the income being dividend income rather than interest income, if more than 40% of the income was made up of these dividends (or the swap was considered a non-qualifying investment). Seems unlikely, but a consideration.

LateGenXer

aroominyork said:

So an ETF like CSH2 should declare all its gains (small 'g') as income, subject to...

LateGenXer said:

Now, how come CSH2 had zero reportable income, I'm not sure, but one theory I see circulated is that the collateral basket of securities it held didn't pay any dividends, or something like that.  It really depends on the financial mechanics of the fund, and what/how HMRC define income vs capital gain.

... before HMRC maybe had a word in their ear?

It's possible there was a change in accounting rules / guidance.

If you check I500 / IE00BMTX1Y45 on https://www.kpmgreportingfunds.co.uk/ you'll see it used to have 0 ERI the first two years, but now has non-zero ERI the last two years.

Anyway, I hope you got an idea of why this ERI is not mechanically identical to the interest yield.  Either way, I think we debated this matter to death now.

masonic

LateGenXer said:

aroominyork said:

So an ETF like CSH2 should declare all its gains (small 'g') as income, subject to...

LateGenXer said:

Now, how come CSH2 had zero reportable income, I'm not sure, but one theory I see circulated is that the collateral basket of securities it held didn't pay any dividends, or something like that.  It really depends on the financial mechanics of the fund, and what/how HMRC define income vs capital gain.

... before HMRC maybe had a word in their ear?

It's possible there was a change in accounting rules / guidance.

If you check I500 / IE00BMTX1Y45 on https://www.kpmgreportingfunds.co.uk/ you'll see it used to have 0 ERI the first two years, but now has non-zero ERI the last two years.

Anyway, I hope you got an idea of why this ERI is not mechanically identical to the interest yield.  Either way, I think we debated this matter to death now.

Well isn't that interesting. Another fund changing its behaviour at around the same time. Certainly supports the theory HMRC didn't like the way funds were interpreting the rules.

LateGenXer

masonic said:

We don't know the details of the swap, so we don't know whether or not it includes dividends from the basket of securities, so it may be that those can make up some of the returns paid to investors. Perhaps this would be covered in the annual report.  [...]  However, if this was the case, that could be an argument for the income being dividend income rather than interest income, if more than 40% of the income was made up of these dividends (or the swap was considered a non-qualifying investment). Seems unlikely, but a consideration.

To be a "bond fund", it needs 60% of the assets in bond-like assets.  Not 60% of the income generated.

Imagine an ETF with 50% Berkshire (zero dividends), and 50% in a high coupon gilt.  100% of the ERI would be interest (as opposed to dividend.)  But it would not be considered a bond fund regardless, as less than 60% in bond-like assets.

masonic

LateGenXer said:

masonic said:

We don't know the details of the swap, so we don't know whether or not it includes dividends from the basket of securities, so it may be that those can make up some of the returns paid to investors. Perhaps this would be covered in the annual report.  [...]  However, if this was the case, that could be an argument for the income being dividend income rather than interest income, if more than 40% of the income was made up of these dividends (or the swap was considered a non-qualifying investment). Seems unlikely, but a consideration.

To be a "bond fund", it needs 60% of the assets in bond-like assets.  Not 60% of the income generated.

Imagine an ETF with 50% Berkshire (zero dividends), and 50% in a high coupon gilt.  100% of the ERI would be interest (as opposed to dividend.)  But it would not be considered a bond fund regardless, as less than 60% in bond-like assets.

Yes of course. Brain is failing me at the end of a long week.

So then if you believe the collateral basket should be included, you'd be forced to conclude all income was dividend income.

aroominyork

I've heard back from Amundi who say:

"As the ETF is an equity fund, it should be taxed under “dividends”. 

Regarding the timing of the fund’s distributions, there will be no separate ERI report provided for the pre-merger period. This pre-merger period will be included in the report which will be due with the new fiscal year end (last day business day of September).

I have attached above the notification that was sent to your custodian regarding the fund’s merger. It is their responsibility to then notify their clients of any corporate actions occurring within a client’s portfolio or holdings."

I cannot see how to attach a file here (can I?) but the extract about the reporting date is (pre-merger on the left, post-merger on the right):

So... would anyone not declare the ERI as dividend?

masonic

I don't think I would because I think their response is a non-sequitur, as it doesn't consider the subject matter of the derivative held by the fund. I suspect they have this just as wrong as they did when they considered it a capital gain.

All of this is reaffirming the wisdom of my decision not to hold such an instrument unwrapped.

aroominyork

The fact that their response doesn't mention the nature of the derivative doesn't mean they haven't considered it. They might be giving a bottom line answer without wanting to open a 100+ comment stream of discussion. 

HMRC's advice to me was to ask the fund manager and, if there remained doubt (which perhaps there is) to add a note on the self assessment. If one's aim is to do all reasonably possible to complete an accurate self assessment, rather than - as it seems is your approach - to do everything possible to avoid underpaying, then declaring ERI as dividend is surely the correct course of action. And anyway, to my simplistic mind, it seems logical that if the returns are liable for CGT rather than income tax, which is the consensus, there is logic to ERI being taxed as dividend. 

If you accept the dividend answer, and given that CSH2 now appears to be declaring the vast majority of its gains as ERI, it once again makes it an attractive unwrapped vehicle. Instead of 18% CGT, it is essentially taxed at 8.75% dividend rate.

masonic

My priority is always to be as accurate as possible, but when I cannot know for certain, to err on the side of caution. Most importantly I avoid actions or inactions that could come back and bite me.

In this case I'm about 80% confident it should be declared as interest because of what the legislation says.

At least the interest on an underpayment now accrues at a more reasonable rate than the previous 8%. Suspect it would be unlikely you'd get a fine as well if you provide the evidence you have. But a wider review of your affairs is likely if one error is found.

Think yourself lucky you aren't held to the ministerial code.

aroominyork

I wonder if we can coax LateGenXer back to give a view...