Potential downsides of an Index-Linked bond ladder?

[MikeT]

Hello, I've read a few of the (excellent) threads on here about nominal and IL gilts and I've learned a lot from them, along with the links to other discussions (Monevator, etc). With IL gilts having positive yields across all maturities they seem (to me) a virtually risk-free way of ensuring the value of today's money in to the future.

Whilst I can see many benefits, I'd like to try and make sure I understand as many of the risks/downsides as possible.


From what I've read it seems some of the main considerations are:

Somebody outliving their ladder

Expenses not matching the income

RPI/CPI not being relevant to personal circumstances



I'm reasonably comfortable with these in as much that:

I'd look to extend the bond ladder out to around age 80 at which point I would use the remainder of my SIPP to buy an annuity

I've gone back through my total annual spending over the last few years and the plan would be to buy each gilt to be slightly in excess of this

I'm not sure there is much I can do about CPI/RPI not exactly matching spending in retirement



Are there other issues I should be thinking about?

Thanks.

[air04]

I have had similar thinking.

My plan is to get ILGs till 90. When one of us(me or my wife) dies or one of us is very ill and we are over 70, I plan to sell the ILGs and buy a joint index linked annuity for the planned amount of income. The ILGs should cover for index linked annuity irrespective of the interest rates etc at that time, as the risk is high(due to age/health/number of people), and as the annuity companies work on average life expectancy(and their pricing will be based on index linked gilts pricing at that time). So, in this way there is no issue of running out of money as long as my inflation is similar to that of RPI/CPI. That helps with the issue of "Somebody outliving their ladder"

[JohnW]

Are there other issues I should be thinking about?’

Treasury default, and needing to sell before maturity, are the ones I see.

‘ , I plan to sell the ILGs and buy a joint index linked annuity for the planned amount of income.‘

Qualitatively it seems to make sense, but put some numbers in. If you cash in 15 years early, your longest remaining bonds will have a duration of about 12 years. Thus, a 2% rise in interest rates at that time will drop bond value 24%. Will annuities get cheaper enough? Do some more detailed modelling, as I doubt it’s fool proof.

[air04]

‘ , I plan to sell the ILGs and buy a joint index linked annuity for the planned amount of income.‘

Qualitatively it seems to make sense, but put some numbers in. If you cash in 15 years early, your longest remaining bonds will have a duration of about 12 years. Thus, a 2% rise in interest rates at that time will drop bond value 24%. Will annuities get cheaper enough? Do some more detailed modelling, as I doubt it’s fool proof.

My (may be wrong) understanding was that an annuity provider would buy something like ILGs (maturity matching cash flow requirements) based on their model of life expectancy. So, if I decide to buy annuity at 70 and life expectancy is 85 for my situation, the index linked annuity provider would quote me price based on the Index linked ladder for the next 15 years; plus their risk/profit on the top. Otherwise, I would need to have a ladder until the maximum I/wife may live. In our case, we plan to leave all to charity on death. I would prefer giving away as much as possible before death.

The drop would be 24% for the 12 year ILG, but may average to a lot less for the ladder.

I would be happy to hear any views about getting guaranteed inflation linked annuity type investment, in any other way.

[MikeT]

Are there other issues I should be thinking about?’

Treasury default, and needing to sell before maturity, are the ones I see.

Thanks John.


Whilst UK treasury defaulting is a possibility (however remote) I would imagine there would be a bit more to worry about than bond prices in that scenario.

I can’t foresee a situation where I would need to sell before maturity. If I there were I’d certainly be wary of buying way out in to the long dates.

[dealtn]

I'd like to try and make sure I understand as many of the risks/downsides as possible.


From what I've read it seems some of the main considerations are:

Somebody outliving their ladder

Expenses not matching the income

RPI/CPI not being relevant to personal circumstances

Consider what happens when real yields turn negative, a position common for much of the last 25 years (a long time relative to a period of normal retirement).

If your ladder is constantly maturing at a real price of 100, but you are continuing to roll and buying at a real price of 120, you are losing real capital, and real purchasing power throughout the life.

[MikeT]

Consider what happens when real yields turn negative, a position common for much of the last 25 years (a long time relative to a period of normal retirement).

If your ladder is constantly maturing at a real price of 100, but you are continuing to roll and buying at a real price of 120, you are losing real capital, and real purchasing power throughout the life.

I was thinking more of buying individual gilts for each year (where available) and holding to maturity. They would all be bought over the next couple of weeks.

I wasn’t thinking of a rolling ladder.

[dealtn]

Consider what happens when real yields turn negative, a position common for much of the last 25 years (a long time relative to a period of normal retirement).

If your ladder is constantly maturing at a real price of 100, but you are continuing to roll and buying at a real price of 120, you are losing real capital, and real purchasing power throughout the life.

I was thinking more of buying individual gilts for each year (where available) and holding to maturity. They would all be bought over the next couple of weeks.

I wasn’t thinking of a rolling ladder.

In which case your ladder outliving you is your biggest risk (but applies to other alternatives too).

[Alaric]

My (may be wrong) understanding was that an annuity provider would buy something like ILGs (maturity matching cash flow requirements) based on their model of life expectancy. So, if I decide to buy annuity at 70 and life expectancy is 85 for my situation, the index linked annuity provider would quote me price based on the Index linked ladder for the next 15 years; plus their risk/profit on the top.

The more likely model they use would be where they work on the basis of having a statistically sufficently large number of lives of the same age. They then project forward the annuity outgo each future year. If they measured the average age at death, that's 85, but their liability for payments and hence demand for matching securities extends well beyond that for those who live into their nineties.

[air04]

My (may be wrong) understanding was that an annuity provider would buy something like ILGs (maturity matching cash flow requirements) based on their model of life expectancy. So, if I decide to buy annuity at 70 and life expectancy is 85 for my situation, the index linked annuity provider would quote me price based on the Index linked ladder for the next 15 years; plus their risk/profit on the top.

The more likely model they use would be where they work on the basis of having a statistically sufficently large number of lives of the same age. They then project forward the annuity outgo each future year. If they measured the average age at death, that's 85, but their liability for payments and hence demand for matching securities extends well beyond that for those who live into their nineties.

I really appreciate your reply Alaric.

When I said "85 for my situation", I meant "on the basis of having a statistically sufficient large number of lives of the same age and health and other risk factors".

Yes, their liability will extend well beyond, but they would have saved money with those who died earlier. That is the risk they have(for which I will pay some extra), which is smoothened by the pool of people they provide annuity. Their liabilities would be peaking at 85 and some form of bell curve I guess around it. They can do it that way as they have a pool of multiple people. As I have only two people(me and my wife), that cannot be smoothened in that way...

Well, that is the way I thought when I came up with doing this with my investments.

[air04]

In which case your ladder outliving you is your biggest risk (but applies to other alternatives too).

Thanks dealtn, that is the reason I was planning to reduce the risk with index linked annuity... bought when one of us dies or is very ill.

[MikeT]

Another question…

For ease of point if we take TR40 with a real yield of 1%.

And we would like £10,000 in todays money in 2040.

Is there a simple formula that would give the required investment today to return the £10,000? Taking 16% off is going to leave me short, it seems more like 13% (£8,700)?

[mc2fool]

Another question…

For ease of point if we take TR40 with a real yield of 1%.

And we would like £10,000 in todays money in 2040.

Is there a simple formula that would give the required investment today to return the £10,000? Taking 16% off is going to leave me short, it seems more like 13% (£8,700)?

The stock answer would be the Present Value (PV) calculation, https://www.calculator.net/present-value-calculator.html.

However, PV assume compound interest, and no tax, and TR40 has a (nominal) 0.625% coupon which, if unsheltered, will be taxed and there is no guarantee that you'll be able to reinvest it at the same rate in order to get the compounding.

TR40 is currently selling for a nominal 95.49 (LSE offer price at time of writing; you may be able to do better) so easiest would be to say that if you want a "real" £10K in 2040 then buy £9,549 of it now, and treat the coupons along the way as beer money.

[air04]

Another question…

For ease of point if we take TR40 with a real yield of 1%.

And we would like £10,000 in todays money in 2040.

Is there a simple formula that would give the required investment today to return the £10,000? Taking 16% off is going to leave me short, it seems more like 13% (£8,700)?

This is how I calculated... There are generally two prices (one with the index linked lift and one without for ease of selling or buying calculation)
from https://www.yieldgimp.com/index-linked-gilt-yields , the price of 94.63 gives real yield of 0.99%. So, if you can buy at 94.63[real price you pay per bond would be 94.63*InflationSinceInception(1.75363)] today, you will need to invest £9463+charges to get £10000 on 22-03-2040. But on the way it will be giving 0.625%(62.5£+inflation) yearly.

Some time ago, here, there were some threads about inflation linked bonds and bond ladders. And there were links to some nice spreadsheets to help to create them. I would suggest you read them.

Some time ago, I created https://docs.google.com/spreadsheets/d/ ... sp=sharing for my use and calculations. That was before I came across https://www.yieldgimp.com/index-linked-gilt-yields

Using my spreadsheet I have created cash flow per year and in that way I know what I get/need.

[MikeT]

I wasn't even close!

Thanks.

[GoSeigen]

I wasn't even close!

Thanks.

You were correct.

If we don't ignore most of the cashflows ("pocket money").

GS

[air04]

I wasn't even close!

Thanks.

The calculation is different as it pays out each year.

If a bond had 1% YTM, and has 15.93 years to maturity(like TR40), and it did not pay any interest until maturity, the price would be 85.33
=100/((1+1%)^[years to maturity]) ... that is close. As there was around 10% in total interest payments, the number you got was different.

In that case you would need to invest £8533 for a inflation adjusted £10000 in 2040. As far as I know all ILGs pay interest...

[GeoffF100]

A big issue here is a change in the tax treatment. If gilts were no longer capital gains tax free that would be a problem. The government has already changed the basis of indexation from 2030. Annuities are guaranteed up to any amount, and the provider takes to hit if the rules relating to index linked gilts change. It would be politically very difficult to plunder people's pensions in payment, but plundering their investments is another matter. On the other hand, annuities appear to have substantial mark ups over the cost of providing them. (I found a mortality table and did a calculation about ten years ago.)

[air04]

A big issue here is a change in the tax treatment. If gilts were no longer capital gains tax free that would be a problem. The government has already changed the basis of indexation from 2030. Annuities are guaranteed up to any amount, and the provider takes to hit if the rules relating to index linked gilts change. It would be politically very difficult to plunder people's pensions in payment, but plundering their investments is another matter. On the other hand, annuities appear to have substantial mark ups over the cost of providing them. (I found a mortality table and did a calculation about ten years ago.)

For me, all money is in tax free shelters, so CGT is not an issue.

Do you have any rough memory of a rough figure of the "substantial mark up" of annuities. I probably could pay 30% extra for the peace of mind.

[Alaric]

Do you have any rough memory of a rough figure of the "substantial mark up" of annuities. I probably could pay 30% extra for the peace of mind.

There are lots of mortality tables and with improving longevity they can become obsolete fairly quickly. Improved longevity reduces the margins on annuities.