Defining the Benefit. When to take a Defined Benefit pension.

Timing risk and when to draw a Defined Contribution pension are well documented. I was fascinated to read RIT's recent post on this in which talks about SWR and links to a video which uses historical data to illustrate that when a pension "pot" is put into drawdown is a major factor in determining whether or not it will last and that no withdrawal rate (whether 4%, or even less) can be regarded as "absolutely" safe but must be assessed in the context of the "value" of the markets at the time the pension is taken.

I was fascinated, but in a detached kind of way, because this kind of deliberation about "when" has never been considered necessary for those of us lucky enough to have index-linked DB pensions. Received wisdom is simple - never take a DB pension before scheme payment age if at all possible, actuarial reduction is to be avoided at all costs. But I've been thinking about this recently and have come to the conclusion that deciding when to take a DB pension is not that simple after all. We may think that it is but that is only because, unlike with a DC pension, it is easy to count the cost of taking the pension early, when what we should actually be doing is making some effort to measure this cost against the benefit.

As an example my own figures come out like this: (I currently have £35,000 in my SIPP and was intending to boost this up to around £50,000, retire at 58 and defer my LGPS till 65).

• Pension if I take it at 65 - £9,300. Tax free lump sum -  £13,000. (When taken this this would be partly subject to 20% tax as I have a small amount of rental income, plus state pension would become payable at 66).

• Pension if I take it at 60 - £7020. Tax free lump sum - £11,500). (This would be taken tax free until 66 as I intend to pay myself just enough out of my SIPP to take me up to the PA).

As I would get the pension for 5 years longer if I take it at 60, I will start to go into a "loss" at age 75 when I will be £2,300 pa worse off. I would therefore be down by around £35,000 if I live to 90 and this loss would increase year on year. Sounds like a bad deal and I should not even be considering it as an option if I can avoid it?

But what this calculation doesn't take into account are the benefits attached to:

• Not having to stretch our finances to allow me to retire at 58 (in other words, the pressure is off as I already have enough in my SIPP. In fact I shouldn't put any more in there as I am already at the limit of what I can use tax efficiently should I decide to access my LGPS at 60)

• Being able to take advantage of a tax free lump sum of £8,500 from my SIPP at 58 which could be re-invested in my ISA, in whole or in part. The rest of my SIPP would adequately fund the two years before I taking my LGPS.

• Being able to access my LGPS tax free lump sum and AVCs at 60 instead of 65 (when my husband would be 71 and we may not be able to put it to such good use). My TFLS/AVC fund currently stands at around £16,000 but could be bumped up to £25,000 by paying what I was going to put into my SIPP for the next two years into my AVC instead. (It is a perk of the LGPS pre-2014 that the whole of the AVC fund can be used to boost the tax free lump sum - subject to certain upper limits that don't apply to me).

• A big part of my income between 60 and 65 would be index-linked and risk free (via the LGPS) rather than managed myself via my SIPP (and therefore subject to market risk or inflation risk if I move it down into cash).

All the above add up to a clear win, for me, to taking my pension early despite the 24% reduction. This win is personal and depends on my lifestyle and situation, what it actually costs in monetary terms is just one of the considerations. After thinking it all through I'm pretty sure which way to go and have revised my targets accordingly. In fact the only one that I haven't yet hit is the one that means I need to add another £1,000 pa to my LGPS pension and that is simple to satisfy  - I just need to keep working for another 2 years.  

On a more general note, if things stay as they are with DB pensions and public sector ones in particular - which is unlikely but does provide food for thought - then it seems probable that some sort of "retirement age" gap could grow up between those with DC pensions, who let the decision on when they want to retire drive their saving and investing plans because no-one can actually tell them in advance what will be the best time to go, and those on DB pensions who just "expect" to have to stay in work until they reach scheme retirement age (which is increasingly being brought in line with State Retirement Age). Of course, there is nothing to stop DB pensioners funding (slightly) early retirement but it does need the foresight to set up an additional personal pension, a fair excess of salary over needs and a willingness to confront the horned beast of actuarial reduction, and assess the benefits of taking a hit on total pension received, in the context of the whole retirement plan, rather than with "just don't do it" blinkers on.

Knowing the financial cost of something is an advantage, but it can be a brake in just the same way as not knowing can (and maybe even more so). When deciding when to take a pension we should all make sure we take due diligence with our cost/benefit analysis and never forget that the only thing we can really be sure of is the value of time.