The GAD numbers that scream for annuity changes

In October 2010 I wrote two blogs about the jiggery-pokery employed by the Treasury to convince us that the reduction in the annual and lifetime allowances would save Britain billions in tax. It wouldn’t , hasn’t and won’t but that’s not the point.

The point of this blog is in the reference made in Mark Hoban’s statement to assumptions published by the Government’s Actuary’s Department in this paper (now archived but available on this link).

In the GAD paper, Trevor Llanwarne who was then and is now the Government Actuary states that he is changing the annuity conversion factors he uses to reflect improvements in longevity and in the discount rate (e.g. the forecast for long-term interest rates).

All pretty archaic you might say, but not a bit of it, these factors should hold good for the next 5-10 years says Llanwarne and they have.

GAD state that the value of a full linked annuity at 60 paid from an occupational pension can be put at 23.6;1 . So your DB pension of £10,000 pa could be valued at 60 at £236,000.

GAD goes on to state that the equivalent cost of purchasing a pension on the open market using an individual guaranteed annuity would be £321,000. (using a factor of 32:1)

That’s a discount of 25% to the occupational scheme or put another way, the occupational is a third more efficient in the payment of its pension.

Now pensions are liabilities and fall on someone’s balance sheet. If they fall on an insurer’s balance sheet, GAD are saying they are considerably more expensive. If an occupational pension scheme wanted to “buy-out” your £10,000 pa pension, it would have to cough up an extra £85,000 to get you off its hands and make you and your life expectancy the insurer’s problem.

So what extra security would you get from being insured by say Legal & General rather than Shell or BP? Well if L&G went bust, you’d have rather fuller protection than if the sponsoring employer went to the wall and Shell and BP might be considered riskier enterprises than L&G (in practice they have very similar ratings but you can imagine them diverging).

But is that greater degree of certainty really worth an increase of £85,000 or a third of the cost of your occupational pension?

Now we can argue all night about GAD’s assumptions, but it’s unlikely that you’d get better data or a more prudent methodology than that employed by Llanwarne and his department. These rates were designed for 5-10 years and none of the fundamentals in terms of the interest rates or life expectancy have changed since then

So we have to assume that the differences in the cost of the pension payable is down to the differing discount factors being used by occupational pension schemes and insurers (both or which GAD approves).

There seem to me (and I’m not an actuary) two factors that can make a difference to the discount factor- investment returns and expenses. Being mutuals and not for profit , occupational schemes do not have to factor in a payment to shareholders, they should have a lower cost base not having to market themselves against competition and they should be able to work on their own longevity data  and be able to make rather more accurate assumptions on the duration of their liabilities.

These are of course the arguments that are made in favour of collective decumulation by the fire-starters who want to move the default pension from the individual annuity basis to the collective basis (as per the occupational scheme).

They have as evidence of the practicality of moving to a collective system, the Netherlands, Denmark, Sweden- even our own States of Jersey, all of which run collective DC schemes which work on considerably lower pension conversion rates than those suffered in the UK by those purchasing individual annuities.

But against the argument put forward by the incendiarists (of which I am proud to be one), are those of the insurers. The ABI have issues 10 good reasons not to abandon annuities. The cynical might collapse them into one “vested self-interest”.

The insurers have a lot to gain from the destruction of the collective system, not least they stand to pick up the assets as they buy-out into insured annuities.

The insurers have much to gain from the replacement system where they not only manage the accumulation, but insure the decumulation through individual annuities.

But a return to collective decumulation would not benefit the insurers, it would mean much of the in retirement wealth of this nation was more directly invested and managed by trustees. Some collective schemes might be insured but , it would be safe to say, some wouldn’t.

So expect to see a number of smokescreens over the next few months

Expect to hear about the added risk of allowing collective schemes to insure their own longevity in mutual pools.

Expect to hear shocking stories of market crashes leaving widows penniless

Expect to hear of the demise of with-profits because assumptions made on equities are always wrong.

All these and more will be coming to a discussion board near you. But be clear about this. Underpinning this argument is governance and the governance we must trust does not come from the insurers but from the Independent Government Actuary, who decides on these things immune from vested interests.

In Trevor Llanwarne’s view, pensions paid from collective arrangements (scheme pensions) are a third more efficient than the equivalent guaranteed annuity.

The argument comes down to risk, are collective DC arrangements a third more risky than individual annuities?

Those are the terms of the debate which we should be having in this country. Forget all the rest.

About henry tapper

Founder of the Pension PlayPen,, partner of Stella, father of Olly . I am the Pension Plowman
This entry was posted in pensions and tagged . Bookmark the permalink.

1 Response to The GAD numbers that scream for annuity changes

  1. Pingback: Can we make the annuity market work? | ToUChstone blog: A public policy blog from the TUC

Leave a Reply