Help on fund charges

[Alaric]

Here's a worked example:

If charges are a percentage of fund value, the higher the fund value, the higher the charges. It doesn't make an enormous difference what you choose as a growth rate. The mathematics remains that if you get a 7% growth with a 1% charge, that outperforms a 4% growth with no charge. If you consider a range of outcomes, the ones where you lose money are considered more likely to come about with the higher growth assets.

Perhaps like gambling, if you went to an infinite number of race courses and always backed at 10-1 or higher, you might end very rich or very poor. If you always backed the favourite, then assuming the reliability of bookies' odds you just end up losing the bookies profit margin.

[SalvorHardin]

Not so. Here's a worked example:

Overall growth at 5%pa over 50 years with no charges: 1.05^50 = 11.467399785753676034851413551211
Overall growth at 5%pa over 50 years with 0.72%pa charges: 1.0428^50 = 8.1292620758362740069967608049655
Factor = 0.70890195054815485320669794735049

Overall growth at 10%pa over 50 years with no charges: 1.1^50 = 117.39085287969531650666649599036
Overall growth at 10%pa over 50 years with 0.72%pa charges: 1.0928^50 = 84.534563963719246164694067941133
Factor = 0.72011201801516932986043374681189

The growth rate net of fees isn't 1 + gross rate - charges. It's (1 + gross rate) / ( 1 - charges)

This changes the maths slightly. So when calculating the return after fees, the equation for value after so many years is:

Starting value x ( (1 + gross) / ( 1 - charges) ) ^ years

You can rearrange this to get separate multipliers for growth and charges, i.e.

Starting value x (1 + gross) ^ years x 1 / ( ( 1 - charges) ] ^ years )

I.e. multiple the starting value by the growth factor and then by the charges factor. So the multiplier showing the effect of the fees is independent of the growth rate.

[mc2fool]

The growth rate net of fees isn't 1 + gross rate - charges. It's (1 + gross rate) / ( 1 - charges)

So with a gross rate of, say, 5% and charges of, say, 0.72%, it would be: (1 + 0.05) / (1 - 0.0072) = 1.05/0.9928 = 1.0576?

So that gives a growth rate net of fees higher than the gross rate before fees. Doesn't sound right to me

[SalvorHardin]

The growth rate net of fees isn't 1 + gross rate - charges. It's (1 + gross rate) / ( 1 - charges)

So with a gross rate of, say, 5% and charges of, say, 0.72%, it would be: (1 + 0.05) / (1 - 0.0072) = 1.05/0.9928 = 1.0576?

So that gives a growth rate net of fees higher than the gross rate before fees. Doesn't sound right to me

Sorry. My fault. That'll teach me to cut and paste on a phone.

The fees term should be 1 + fees, not 1 - fees. The conclusion still holds

I'm a bit rusty, years spent not doing much maths have clearly taken their toll. Watching Jose Mourinho's latest press conference at the same time as typing probably had something to do with it

[mc2fool]

The fees term should be 1 + fees, not 1 - fees. The conclusion still holds

Arghhhh!!! You're right, and especially "Arghhhh" as I have often pointed out exactly the same error to folks on these boards subtracting inflation (rather than dividing) to get real returns!

To wrap up though, GeoffF100's "factor of (1 - 0.72 / 100)^50 = 0.696768" (and your post agreeing) also isn't quite correct, for the same reason.

It should be (1 / (1 + (0.72 /100)))^50 = (1/1.0072)^50 = 0.698577... not a huge difference I agree, but still ...

[SalvorHardin]

Arghhhh!!! You're right, and especially "Arghhhh" as I have often pointed out exactly the same error to folks on these boards subtracting inflation (rather than dividing) to get real returns!

To wrap up though, GeoffF100's "factor of (1 - 0.72 / 100)^50 = 0.696768" (and your post agreeing) also isn't quite correct, for the same reason.

It should be (1 / (1 + (0.72 /100)))^50 = (1/1.0072)^50 = 0.698577... not a huge difference I agree, but still ...

Agreed. It's not my day too, so double aarrrgh and a Homer Simpsonesque "DOH!" as well

Change the fees term in my post to ... x (1 - charges) so the multiplying factor for charges (instead of a divisor) becomes (1 - charges) ^ years

Apologies to everyone. The conclusion still holds. To think I used to be an Actuary (albeit 15 years ago). Mourinho's press conference has a lot to answer for, as does the Ryder Cup (the only time I seriously watch golf)

[mc2fool]

Arghhhh!!! You're right, and especially "Arghhhh" as I have often pointed out exactly the same error to folks on these boards subtracting inflation (rather than dividing) to get real returns!

To wrap up though, GeoffF100's "factor of (1 - 0.72 / 100)^50 = 0.696768" (and your post agreeing) also isn't quite correct, for the same reason.

It should be (1 / (1 + (0.72 /100)))^50 = (1/1.0072)^50 = 0.698577... not a huge difference I agree, but still ...

Agreed. It's not my day too, so double aarrrgh and a Homer Simpsonesque "DOH!" as well

Change the fees term in my post to ... x (1 - charges) so the multiplying factor for charges (instead of a divisor) becomes (1 - charges) ^ years

That looks even better* -- and means GeoffF100's original is correct -- so I'm not sure what you're saying "Agreed" to!

* I often find issues with formulae come to light if you stick in silly numbers, e.g. fees of 50%!

[GeoffF100]

The fees term should be 1 + fees, not 1 - fees.

That is not quite right. If you get g% growth in year 1, the value of your fund at the end of the year, before costs are deducted is the original value multiplied by 1 + g / 100. Fees are then deducted. The fund manager takes c% of your original value multiplied by 1 + g / 100, i.e. your original value multiplied by (1 + g / 100) * c / 100. That leaves you with your original value multiplied by (1 + g / 100) * (1 - c / 100). after n years this becomes your original value multiplied by (1 + g / 100)^n * (1 - c / 100)^n.

As an aside, if x is small, 1 - x is approximately 1 / ( 1 -x) = 1 + x + x^2 + x^3 + ....

[Alaric]

That is not quite right. If you get g% growth in year 1, the value of your fund at the end of the year, before costs are deducted is the original value multiplied by 1 + g / 100. Fees are then deducted.

I think you will find that in an OEIC where the fees are included in the price, that they will be accrued on a daily basis and physically removed at a later date. Any other method in an open fund overcharges entrants during the year and under charges those leaving.

Platform fees on the other hand will be calculated according to the rules of the platform and there's no particular reason why they wouldn't choose to levy them on period end values. There again they might accrue them daily. Anyone with any practical examples?

[TheMotorcycleBoy]

As always with percentages, one should be clear what it is a percentage of. When you mention the iweb reinvestment charge, is that 0.5% of the amount reinvested? If so then your preference share return is 0.995 times what the running yield is quoted as.

That's correct, Alaric, so if we've got £1000 of holding, get a £10 div, and it all were to be reinvested, we get charged 5p.

What you are seeing is an illustration of the trade off between risk and return.

The 4% fund has a lower return but less credit risk. At the cost of higher credit risk and higher charges, you may get a higher return at the "Opportunities" fund. But it all might go wrong and the 4% fund could ultimately have been the better investment.

Yes, that's what I thought.

If you are investing in a tax free fund, a good general purpose bond fund is the Vanguard Global Bond Fund (hedged to Sterling). That spreads your risk over a huge number of international bonds with low to medium risk.

Thanks Geoff, but we really do want a high risk/return fund. We have high yielding IPF1 and PMO1 already, and wanted to purchase some other high yielders e.g. Coops 11% 2025 (I forgot the code), maybe EROS but it now seems that our ISA platform iWeb doesn't do a good choice of retail bonds, so like I wrote here, that's why we are looking for a high yield bond fund.

BTW, many thanks for all the replies, I'll read them all and get back, but to be honest, I need to spend the remainder of the day doing something not figures or PC related!

Good evening to you all

[Alaric]

That's correct, Alaric, so if we've got £1000 of holding, get a £10 div, and it all were to be reinvested, we get charged 5p.

That seems an unusually cheap price. Is there perhaps a minimum charge somewhere in the small print? A minimum amount that can be reinvested even.

[GeoffF100]

[I think you will find that in an OEIC where the fees are included in the price, that they will be accrued on a daily basis and physically removed at a later date. Any other method in an open fund overcharges entrants during the year and under charges those leaving.

If fees are accrued at vanishingly small increments of time (which is a good approximation for daily accrual) we get an exponential:

http://www-stat.wharton.upenn.edu/~wate ... node13.htm

This does not, however, change my result.

[GeoffF100]

Thanks Geoff, but we really do want a high risk/return fund. We have high yielding IPF1 and PMO1 already, and wanted to purchase some other high yielders e.g. Coops 11% 2025 (I forgot the code), maybe EROS but it now seems that our ISA platform iWeb doesn't do a good choice of retail bonds, so like I wrote here, that's why we are looking for a high yield bond fund.)

If you want risk, you could consider:

https://www.vanguardinvestor.co.uk/inve ... _fund_link

However, high risk does not necessarily mean high return. It means you are likely to lose your money.

[hiriskpaul]

High yield spreads and default rates are very low, nearly as low as just before the financial crisis, when they went sky high. I take this as a sign of caution rather than as a sign of great value in high yield bonds. HY is unlikely to hold its value when the next stock market slump comes along either, but if you really want to put money into high yield, take a look at iShares Global High Yield Corp Bond (GHYS).

GHYS is GBP hedged, so fluctuations in the exchange rate are eliminated (mostly) and so will act as though they were a portfolio of GBP denominated bonds, but with yields adjusted for the differences in the various currency curves. Weighted average yield to maturity is 4.91%, but you should treat that only as a very rough guide to future expected return. They can make more over time as some credit ratings will rise and they sell for a profit, but on the other side there will be some defaults resulting in partial or total losses for some bonds. There is also the fund management charge of 0.55% and costs of buying/selling bonds to come out, although these transaction costs are unlikley to be very high, around 0.1% as an educated guess.

GHYS is the best you will do for a high yield passive fund with iWeb. The US market is much better for high yield bond ETFs with much lower charges and more variety, but US listed ETFs off limits to you as of the start of this year.

I see GeoffF100 has mentioned VEMT. Good low management charge for an EM bond fund, but not currently available at iWeb. If you decide you do not want currency hedging, iShares SHYU is available, which is a very large ETF covering the USD high yield market. There is a GBP hedged version of that which I would recommend over GHYS, but not available at iWeb!

[hiriskpaul]

However, high risk does not necessarily mean high return. It means you are likely to lose your money.

Well to be pedantic, as seems to have happened on this thread, it means you are more likely to lose your money. It is a very good point though. In the last financial crisis the USD high yield sector lost around 30% of its value (peak to trough) compared with only 10% for investment grade. US Treasuries went up. Ok, high yield did not drop as much as equities, but it did not manage anything like the subsequent longer term returns either. I personally think a better strategy is to ignore HY completely, unless you pick your own, and go for a combination of investment grade (or government bonds) and equities. Ongoing running costs tend to be much lower for investment grade and equities as well.

[GeoffF100]

My conclusion has been to ignore HY bonds completely. If you can accept more risk in the hope of higher returns, you can increase your proportion of equities. You cannot build a well diversified portfolio of exclusively UK HY bonds, but that does not matter too much if you already have a well diversified equity portfolio. Nonetheless, you are betting that you know more than the market, which generally is not a wise thing to do.

[TheMotorcycleBoy]

Firstly thank you all very much for diving in with the fee/growth calculations. And my apologies to Salvor for disturbing his Sport updates! To be honest the maths part does not challenge me a great deal, since it was a strong point in my education and I still apply myself these days. (And FWIW I agree with Alaric when he said something about the charge being applied daily against the fund's value at close of day. After my convos of late with pension providers that's usually what they do.)

But what I most appreciate/enjoy is listening to you people and hearing what is the experience that Mel and I lack. And I think I've got a rough idea of what you (all?) are saying:

However, high risk does not necessarily mean high return. It means you are likely to lose your money.

High yield spreads and default rates are very low, nearly as low as just before the financial crisis, when they went sky high. I take this as a sign of caution rather than as a sign of great value in high yield bonds. HY is unlikely to hold its value when the next stock market slump comes along either, but if you really want to put money into high yield, take a look at iShares Global High Yield Corp Bond (GHYS).

My conclusion has been to ignore HY bonds completely. If you can accept more risk in the hope of higher returns, you can increase your proportion of equities. You cannot build a well diversified portfolio of exclusively UK HY bonds, but that does not matter too much if you already have a well diversified equity portfolio. Nonetheless, you are betting that you know more than the market, which generally is not a wise thing to do.

I *think* that I'm starting to understand what you lot are trying to imply. Basically I *think* you saying that when times are hard the high yield/risk fund will take a bigger hit then the lower yield/risk one and therefore it's return will fall to a level equivalent to that at which the lower yield one is currently at, but since the high yield one's OCF is greater, the net result for the investor is actually less. Correct?

However, surely, since we are talking about Fixed Income instruments here that would imply that if we assume a high yield bond portfolio containing 100 long-dated securities all with running yields from time of introduction to the fund of 6%, then 33% of it's holdings would need to go into default before it's average running yields fell to 4%, the yield figure for a less risky / lower returning fund?

Is that the kind of the scenario that you all are saying we must be mindful of?

thanks
Matt

[hiriskpaul]

I *think* that I'm starting to understand what you lot are trying to imply. Basically I *think* you saying that when times are hard the high yield/risk fund will take a bigger hit then the lower yield/risk one and therefore it's return will fall to a level equivalent to that at which the lower yield one is currently at, but since the high yield one's OCF is greater, the net result for the investor is actually less. Correct?

However, surely, since we are talking about Fixed Income instruments here that would imply that if we assume a high yield bond portfolio containing 100 long-dated securities all with running yields from time of introduction to the fund of 6%, then 33% of it's holdings would need to go into default before it's average running yields fell to 4%, the yield figure for a less risky / lower returning fund?

Is that the kind of the scenario that you all are saying we must be mindful of?

thanks
Matt

No, you have to consider capital losses as well as non-payment of coupons. Default on 1 bond in 100 could potentially mean a 1% loss on the portfolio. Not all defaults will result in a total wipeout of course, but often bondholders will not recover all their capital.

Running yields are not a very useful concept by the way. For bonds yield to maturity is better. Running yield on Coop group 11% 2025 is 8.5%, but yield to maturity is only 6%. The yield to maturity is better as it takes into account the 30p per bond capital loss from now to maturity.

[Alaric]

The yield to maturity is better as it takes into account the 30p per bond capital loss from now to maturity.

Bond funds seem to be run on the basis of avoiding capital loss from this effect. Trading to keep the duration relatively constant is what they probably do. The risk is to capital value as well as default or partial default, so that a bond with a 6% redemption yield widens out to a 10% yield. How much that writes down the capital value would depend on the outstanding term to maturity.

[GeoffF100]

I *think* that I'm starting to understand what you lot are trying to imply. Basically I *think* you saying that when times are hard the high yield/risk fund will take a bigger hit then the lower yield/risk one and therefore it's return will fall to a level equivalent to that at which the lower yield one is currently at, but since the high yield one's OCF is greater, the net result for the investor is actually less. Correct?

You are on the right lines. In the very long run, one would not expect the return from high yield (AKA junk) bonds to fall all the way down to those of investment grade bonds. We would expect there to be some residual long term reward for holding these bonds. The reason is that they have an unattractive risk/reward profile. Investment grade bonds are likely to deliver a reliable return come what may, except in the most dire of circumstances. Junk bonds give good returns in the good times, but terrible returns in the bad times, just when you can least afford to take the hit. You cannot predict in advance whether times will be better or worse than the degree of badness that is already priced into the bonds. People are usually willing to take a lower return from an investment that delivers reliable returns and pays up when they need it.