# 'Warning! Axa Sun Life and other over-50s plans (but don't just...) blog discussion

edited 2 April 2012 at 6:02PM
in Martin's Blogs & Appearances & MoneySavingExpert in the News

16 replies
5.5K views

Former_MSE_Helen
Former MSE

2.4K Posts

This is the discussion to link on the back of Martin's blog. Please read the blog first, as this discussion follows it.

Please click 'post reply' to discuss below.

**Read Martin's "Warning! Axa Sun Life and other over-50s plans (but don't just cancel if you've got one)" Blog.**

Please click 'post reply' to discuss below.

0

This discussion has been closed.

Latest MSE News and Guides

### Martin Lewis quizzes Rishi Sunak

Watch the cost of living support Q&A here

Join the MSE Forum discussion
## Replies

For Beatrice her cohort life expectancy at 60 in 1997 was 82.4 years. At 75 in 2012 it's 88.1 years. Ignoring investment returns it would still on average make sense for her to continue paying in.

In practice she can put the monthly premiums into an ISA and it'll probably grow by enough to make paying in the wrong choice on average. For it to be the optimal choice to pay in she'd need to be closer to 80 than 75.

This leads to some somewhat reasonable general guidance:

The education, regional and health factors are very crude but enough to give some idea of how things change and produce somewhat decent results, though a more detailed tool could definitely do better. At the extremes:

Man in Scotland who was a labourer, has artery disease, smokes, drinks a lot and is obese, good to continue paying in from age 62.

Woman professional in a rich area with no health trouble, good to continue paying in from age 86.

You probably already know about the difference between period and cohort life expectancies so this is mainly a note for others that the true age to which at least 50% of people their age will live is likely to be a few years longer than shown.

Given your liking for statistics you're probably itching to have a calculator that can work out the likely best choice for each person...

I am glad that you are warning against these nasty little policies, with high charges and frequently inadequate cover. But why the outrage at the facts qoted above?

Surely this is normal for any type of sustainable insurance?

{Payments in} = {Policy admin overheads} + {Insurance company profit} + {Payments out}.

It is only to be expected that, on average, people will pay in more than is paid out.

It's the same with e.g. my house insurance. I expect to keep paying in to cover my risks, but I would very much prefer never to have a payout. Its the same for the majority of buildings policyholders. Only the 'lucky' few have their house burn down and get paid out more than they put in.

Interesting stuff - thank you

Martin Lewis, Money Saving Expert.Please note, answers don't constitute financial advice, it is based on generalised journalistic research. Always ensure any decision is made with regards to your own individual circumstance.

Don't miss out on urgent MoneySaving, get my weekly e-mail at www.moneysavingexpert.com/tips.Debt-Free WannabeeOfficialNerd Club: (Honorary) Members number 000Except this isn't insurance its assurance - as death is assured. My problem here is the very poor return against a definite circumstance.

In effect you're insuring against dieing earlier than average - and that's really how it needs to be portrayed, i simply dont think many people who get the policies perceive it as insurance and especially not insuring against early death.

Martin Lewis, Money Saving Expert.Please note, answers don't constitute financial advice, it is based on generalised journalistic research. Always ensure any decision is made with regards to your own individual circumstance.

Don't miss out on urgent MoneySaving, get my weekly e-mail at www.moneysavingexpert.com/tips.Debt-Free WannabeeOfficialNerd Club: (Honorary) Members number 000The advantage of this approach is that you can have your money sitting in an account and making you some money while you're still alive and yet still have it available very quickly to take care of funeral expenses. Since you need an emergency fund anyway, this can be doing dual duty as both your emergency fund while alive and funeral expenses when needed.

For non-funeral expenses the joint account approach can be a good one, though do be aware that it will create a financial association between the credit records of each joint holder and that may be undesirable in credit reports.

If you want to make life easier for the person handling the immediate things:

1. Write down your funeral preferences.

2. Write down a list of people you want to be told and invited, and say if there are any you don't want invited.

3. Write down a list of your accounts and contact information for them all.

4. Ensure that any executor named in your will is really willing to be your executor. If they refuse things can get very messy.

I see. Thanks for responding Martin.

Possibly a 'fairer' way to sell such policies would be to make them 'paid up' after a time.

For example just making up some figures to illustrate: for the promise of a payout of £1000 upon death, charge a premium of £10 a month, but after 15 years of premiums (£1800 paid in) if the policyholder still survives stop collecting the premiums but stil pay out £1000 when death eventually occurs.

Still a pretty pointless product for the majority of people. The main problem is they are (mis)sold mainly on the emotional benefit of 'caring for loved ones when you are gone' rather than any truly practical benefit.

The majority of policies probably would count as mis-sales if the insurer was responsible for ensuring the product was suitable for a customers needs. But these are sold on the basis of 'guaranteed acceptance', no questions asked (i.e. no fact find).

But that to me is the biggest problem with the article.

You cannot ignore investment returns.

To consider one of the examples in the article:

'

Big Bob is a 65-year-old darts player in pretty decent health. His kids are struggling and he’s alone, so he decides to put a bit of cash aside each month in an over-50s plan, paying £5 a month with Axa. It promises to pay him out a guaranteed £660 when he dies provided he lives at least two years.'

There are a few things that give an indication about Bob's life expectancy here.Big Bob - is he obese, or an alcoholic, or both?

Darts player - likewise, does he spend a lot of time in the pub, or smoking?

He's alone - this will reduce his life expectancy, for a number of reasons

Kids struggling - not sure what this means, but he has some support here, probably increasing his life expectancy.

Anyway, moving on, let's just say Bob is 'average', and can expect to live around 21 years.

Regardless of the value of the final payout, the problem here is that the longer you pay in, the more significant inflation becomes.

To simplify it, let's say there are annual payments of £100 at the beginning of each year, and you get back £1000, and inflation is 5% per year.

If you live 2 years, you pay in £100 in today's money in year 1, £95 in today's money in year 2, therefore paying in £195 of net present value, and you get back £1000*.95*.95 = £902.50 We pay in a nominal £200 and get back a nominal £1000, but we paid in a real £195 - 97.5% of the nominal value, and got back a real £902.50, only 90.25% of the nominal value.

What happens if you were to pay in for 10 years + then die? This is a geometric series, n=10, r=0.95, a=£100, so the NPV of the payments is £802.56, while the NPV of the payout is only £598.74; this time the real value of the payments is 80.26% of the nominal, but the real value of the payout is just 59.87% of nominal.

In other words, while they might quote higher payouts for younger people, by the expected date of death, these payouts will be ravaged by inflation, relative to the amount paid in - ALL of your payout will be in inflated, 2033 (expected date year death) pounds, whereas your payments in will be partly in expensive 2012 pounds, and only the very last payments will be depreciated to the extent of the ENTIRE payout.

Taking the example from the article:

This isn't right at all.Find out how long it’ll be before he pays in more than it’ll pay outDivide the payout by the monthly contribution:

£660 divided by £5 = 132.This gives the number of months after which he will have paid in an amount equal to his lump sum.

132 months divided by 12 = 11 years.So, as Bob is currently 65, if he reaches the age of 76, he will have contributed more than the planned payout.

Let's derive a better formula.

Let (100% - monthly inflation) = r, the number of months premiums paid = n, the assured amount = m, and the monthly payment = a

The NPV of the assured amount is m * r^n.

The question is when does this equal the NPV of the sum of payments

This sum is given by a * (1-r^n) / (1-r)

So in general this becomes a good deal when m * r^n > a * (1-r^n)/(1-r)

If annual inflation is 5%, r = 0.995926, we know m = £660, a = £5, so

660 * 0.995926^n > 5/0.004074124 * (1- 0.995926^n)

i.e. 660 * 0.995926^n>1227.258-1227.258* 0.995926^n

or

0.995926^n > 1227.258/1887.258

i.e. n = log(1227.258/1887.258, 0.995926)

= 105.46 months, or just short of 9 years

Another point from the article

Though it’s important to note that if you were to die during the first one or two years (which varies by policy) you would not get a payout.Not quite true.

Per Axa,

If you die in the first 2 years, we’d pay back all the premiums paid, plus half as much again.A male aged 85 paying £74/month assures £4,055 on death. He has approximately a 22% chance of dying before 2 years, in which case he gets back all the premiums paid, plus 50%.

Here are US mortality rates by age:

http://www.ssa.gov/oact/STATS/table4c6.html

(can't find UK ones)

I plugged these into a spreadsheet and calculated the probability of death at any given age for someone age 85, and the NPV of payments and death benefit at that age.

From this, I calculate the profit if you die during a given year.

For example, if you live to be 105, you will receive the lump sum with NPV of £1381, but you will have paid premiums of £11,440. This gives a loss of NPV from dying at that age of £10,060.

On the other hand, IF you die after 2 years (say 2.5 on average), your profit is £1389, based on having paid 2.5 years of premium and the NPV of the death benefit. Sum all the possible death years and you find the profit has an expected loss of £1612.50

Take the example of a 65-year-old man who’s been told he’s only likely to live for another five years. On that basis, even an Axa plan for £74 a month would pay out £12,085 – if you did die after five years you’d only have paid in £4,400 to get it, a very efficient investment.So for someone who’s very unlikely to make the average life expectancy, these can be a seriously good gamble.Well that really depends on the nature of what's limiting your life expectancy. If you look at the actuarial tables above, mortality risk for a 65-year-old in the next year is 1.7%. In order to have a life expectancy of 5 years, it's likely that your mortality risk in the next year is say 17%. But it might not be, and whether it's a good deal depends on these numbers. If for example the doctor says 'we expect you to live five years, but if you beat that, it's likely to be twenty', then when you combine the fact that you don't get the full payout until 2 years, and for someone with limited life expectancy, then there's presumably a substantial chance that they will die within 2 years, all but killing the payout, with the chance that if you return to health as a 70-year-old, you can very likely face another 20 years of life, it can add up to a poor deal.

Bottom line of course is that these policies are almost always a bad deal.

This depends on the actual payout of your personal policy. Perhaps there are some historical policies that have generous rates, I don't know. In order to make a general assessment, you need:

- monthly payment
- assured amount
- age
- sex

And then you can make health/lifestyle adjustmentsI think you are conflating risk factors and indicators of those risk factors, but the two are not the same.

Someone who is a labourer or Scottish is more likely to make bad lifestyle choices, SUCH AS poor diet (artery disease?), smoking, obesity, excess drinking, and so on. But if you are a Scottish labourer eating healthily, enjoying a decent pension, exercising regularly, etc., then those factors are what count, and your life expectancy will be on the right-hand side of the curve. Being Scottish does not shorten your life expectancy, and you shouldn't adjust based on it for individual cases - it's only useful for considering the cohort as a whole (e.g., do people in Scotland get a worse deal from these schemes than people in England), not for individuals.

ONS probably has some, you might try some searching around their site.

I agree. I was mixing the two deliberately, using some personal factors and some indicators to keep things simple. Should really be done with more in depth individual questioning. Good enough for my purpose - a short post to illustrate a conceptual way of varying things - but definitely imperfect. The personal risk factors aren't really independent of each other either but trying to get into that would have been way over the top...

I did, though only up to age 89 The US figures for 90+ seem to fit so I used those for ages of 90 and above for my spreadsheet:

http://www.!!!!!!/document/3BBzPnmA/actuary.html

These plans have rather more variations than discussed in the article.

For instance the Asda/LV policy has a capped policy, where you won't pay more premiums than you get out.

Now it's a given that these policies are almost always a bad idea, in the first place, but if you've held a capped policy for a long time, it's a better bet to keep it.

E.g., Asda/LV 50+ plan, male age 70 assures £6,034 @ £50 PCM with uncapped premiums, or £4,281 @ £49.34/month with capped premiums

The expected value of these policies (taking into account the risk of dying in year 1) is -£2,668.81 and -£1,156.10 respectively. Both a poor deal, but because the cost of capping (with Asda/LV) is constant with age, the older you are the better a deal it is (since for a younger person the assured amount, by the time of death, will be ravaged by inflation anyway).

After 3 years of payments, however the capped policy is worth £312.52, while the uncapped policy remains resolutely an absolute stinker, with an expected value of -£1583.99

Basically, although an uncapped policy is also a bad deal, unless you are either young (closer to 50 than 70), or just took it out very recently, it's worth keeping. A capped policy - not so much.