The latest fad is unwinding annuities. I am not against it in principle, but you have to be satisfied that it is the right thing to do, and if you do decide to do it, the basis has to be right.
The issues to consider are:
1. Insurers who have issued annuities would have purchased gilts to back them. When these are offloaded onto the market what impact would they have on quantitative easing?
2. Some insurers would have purchased corporate bonds or other assets in the hope of higher yield. These are less marketable and could cause liquidity problems and, in the extreme, damage the solvency of the company.
3. What about annuities – for that is what pensions are – payable from self-administered defined benefit schemes? Who would define the value of the annuities? Would you use a common basis for all pensioners in a scheme or take their health into account?
4. Most schemes are not fully funded. If all of them had to pay out the cash value of pensions that have already commenced, then the pensions of current employees and past employees who have not commenced their pension would be even more thinly funded. Trustees have a duty to all. How would they discharge their responsibilities?
5. If you allow unwinding annuities for private sector defined benefit schemes it would be difficult not to justify it for public sector schemes. The same issues arise for funded public sector schemes – but what about unfunded ones? The cash would have to be found out of the current year’s taxation or by further borrowing.
6. What would all this cash becoming available for investment in a short space of time do to the investment market? Give it an artificial puff?
7. If the government continues to keep interest rates low, people will find that they have to invest in riskier assets to replace the pension foregone. That could have unfortunate consequences.
8. Finally, it is unrealistic to allow unwinding for one class of pensions but not others, as the government might do in response to some of the concerns I have highlighted. Once the genie’s out of the bottle you can not put it back.
In setting the terms for an annuity the crucial assumptions are those of mortality and investment return. If the client wishes to cancel, say, six months after the annuity was purchased, the investment conditions could be different. If the underlying interest rate was 1 per cent a year higher then the investment the insurer is holding to underpin the annuity would be worth perhaps 10 per cent less. Also the insurer would be fearful that you were wishing to surrender due to poor health so he would make a heavier mortality assumption. It may be therefore that you would get back no more than 80 per cent of your monies. Now the so-called experts would say that the insurer has pocketed 20 per cent but that is not the case. It is value-neutral.
Of course it can go the other way – the interest rate could be 1 per cent lower, so the investment the insurer is holding is worth 10 per cent more, which would offset the mortality assumption, so the client could get more money back. He would not ‘win’ because that would not buy a higher income than he had foregone.
I have constructed a simple model to illustrate the issue of mortality. Let us assume that the standard mortality rate at age 65 is x per 100. This is an average for all annuitants, in various states of health. The distribution is very lopsided as overall they need to average out. If 1 per cent of them have thrice the average mortality and 4 per cent twice then as much as 60 per cent would have to be 10 per cent lighter than average. Why?
Simple algebra:
1% of 2x + 4% of x = 0.06x
60% of 0.1x=0.06x
So the heavier mortality and lighter mortality cancel out.
Using my mortality model, the expectation of life for an average 65-year-old is 22 years. Using an interest rate of 3 per cent, an insurance company can offer for a purchase price of £100,000 either a) an annuity certain of £5,904 for 22 years; or b) a lifetime annuity ceasing on death of £6,406 a year.
The lifetime annuity of £6,406 a year is £502 a year higher than the annuity certain of £5,904. Potentially the insurer could be paying it to age 100 or more. How can it afford it? Well, a few will live for fewer than 22 more years. The money saved from that pays for the longer-lived, as well as for the extra £502 a year. The suggestion that the insurance company ‘pockets’ the money on early death is misguided.
If the insurer were to take the annuitant’s health into account, he could be offered either a) £6,973 if the death rate is twice normal; or b) £7,889 if the death rate is thrice normal.
What terms does it then offer to the others? If they were to assume that they were all super-fit, then my calculations indicate that it can only offer £6,224. However, in my example, 60 per cent are super-fit and the remaining 35 per cent (100-60-4-1) per cent should get the average rate of £6,406 a year. In practice no one will volunteer the information that they are super-fit and neither can the insurer insist on the information. So it will quote a rate somewhere between £6,224 and £6,406 – say, £6,250 a year.
They are five different bets as far as the insurer is concerned. They would not expect a change in bet mid-course.
It may be that a company does not offer better terms to impaired lives. That does not matter as long as such annuities are available from another provider. That company would have to assume that its average would be lighter than would otherwise be the case.
If one of these guys comes back and says he now wants an impaired life annuity can he cancel, what should the insurer do? If he is really impaired – say, to thrice normal mortality – the insurer is buying back an annuity on the life of an impaired life. So they will only offer 100,000x6,406/7,889 – that is, £81,202. The insurance company has made neither a profit nor a loss.
If the FCA or the ombudsman says the customer should be offered a full refund then it is the shareholders who lose out. Before you smile and say: “tough”, ultimately it is you and I who lose, as these shares are held by pensions funds and unit trusts.
In summary, I do not object to an annuity being unwound, but it needs to be done on the correct basis reflecting changes in investment and mortality. But first and foremost the government must be satisfied that it is the right thing to do. Otherwise it will cause long-term harm for the sake of short-term electoral gain.
Icki Iqbal is a former director of Deloitte
Key points
* You have to be satisfied that unwinding annuities is the right thing to do – and if you do decide to do it, it has to be on the right basis.
* In setting the terms for an annuity the crucial assumptions are those of mortality and investment return.
* The lifetime annuity of £6,406 a year is £502 a year higher than the annuity certain of £5,904.
