Retirement Strategies 2 – SIPP Run-Off and the LTA
You're trying to reach a page that is behind our paywall.
You can sign up to read it by pressing the Patreon button below - thanks.
Patreon is a membership platform that helps people take out monthly subscriptions to online content they want to support.
You can sign up to read it by pressing the Patreon button below - thanks.
Patreon is a membership platform that helps people take out monthly subscriptions to online content they want to support.
Unlock with Patreon
Really useful Mike, this kind of thing is rarely discussed. With a HRT tax threshold of 46350, my strategy would be to withdraw 61800 (to include lump sum). Then with 2% growth the pot vanishes at 75 with 1297k withdrawn and only brt tax paid. With a 8500 state pension coming in at 67 you have to slow withdrawal and end up with 57k at 75 which has the tax to pay. You’d then withdraw 4405 a year to keep inside personal allowance and the dregs would last 10 more years.
I think you spreadsheet has an an error as the withdrawals are’t reducing the pot fast enough for the first 2 years. You have 1000k, 1020k, 1040k, but the numbers should drop below 1000k in year 2
Thanks for that John – you’re right. The second and third row have the wrong formula. I think it’s because I started with a ral world spreadsheet where I didn’t take any money in the first two years after hitting 55.
I’ve fixed it now, but it doesn’t change the results much.
Thinking about it more. I see you are doing UPFLS to avoid getting the 25% lump sum taxable as you drawdown. If you do go for drawdown and take the 250k in cash, put it in 3% dividend equities, 1% capital growth, inflation 2% (so 2% real return) and trickle it into an ISA, it will take 17 years to clear, in which time you’ll have got 70k in dividends, but only paid tax on 40k of them, and at 7.5% so 3000. You are well under cgt limits.
That doesn’t work for me as the UFPLS maxes out my basic rate tax band so I would pay 32.5% on divs. It’s annoying that the government has started taxing divs so much as they are now very unattractive outside SIPPs and ISAs.
I’m actually avoiding taking the 25% tax-free lump sum because I have nowhere sensible to put it. When my other half retires it’s quite likely that we will each take our TFLS and buy a cottage by the sea.
Mike, have you given any thought as to how DB pension(s) – with their own LTA issues – coming on stream at say age 65, would effect the three approaches you have modelled?
It would be similar to the impact of the State pension, which I’ve added in from age 67 (by reducing the UFPLS). Can’t see how it would affect any of the draw-down schemes more than the others).
I still don’t see why you venture into hrt. If your spending is say 60k, just use the lump sum dividends, add drawdown and state pension up to hrt limit and top up burning the lump sum capital. It will last nearly 2 decades. Then top up by burning ISA capital in the same way. Eventually you might need to suck hard on your pension, more likely it will be a iht free part if your estate.
Mike, I think that you could draw income from your £1m Pot with 0 unavoidable LTA charges if you use drawdown (d/d) rather than UFPLS. This is because it is net growth (investment growth less income payments) that is tested at age 75 for d/d – so called benefit crystallisation event (BCE) 5A. Specifically, provided the Pot is worth less than £750k at age 75 (assuming £250k PCLS taken at commencement, as John B has suggested) there is no LTA to pay. Thus, there is also no pressing need to empty the Pot by age 75 if you go the d/d route. On a like-for-like basis the d/d approach may incur more income tax than UFPLS route but it seems to be better suited to managing the LTA levy – which as you point out cannot be offset.
The reason I asked about DB schemes is that I think any DB pension coming on stream will, unlike the state pension, come with an LTA hit (20* gross pension I believe) and the strategy should ideally consider such upcoming LTA hits too.
You are of course right about the 20x DB hit (a real bargain, as it happens) but there’s no decision-making with DB pensions. The purpose of the post was to compare rundown strategies and I don’t think adding in a DB pension would make the comparison clearer or change the outcome.
I agree with you about drawdown. I have thought about moving into drawdown if the markets fall (shrinking the pot size). Currently, I’m taking my money out of the pension as quickly as I can because I don’t expect the rules to stay as they are, and because I want to use up my basic rate tax band every year.
I guess I’m also hoping at the back of my mind that they will abolish the LTA at some point.
According to Saturdays (18/8/18) Times, the tax take due to the LTA has risen from £10m in 2006/07 when the LTA was introduced (at a level of £1.5m) to £100m in 2016/17. Unfortunately therefore, I cannot see the LTA being abolished any time soon.
I can see that a transition to drawdown could be a beneficial side-effect of a market correction. Furthermore, such a transition could be executed in a number of discrete steps or phases.
I agree that currently the 20* factor seems good value. But perhaps this been tempered, or at least capped, by reducing the LTA limit, from its height of £1.8m in 2010/11 and 2011/2012.
Re DB decision-making, depending on the DB scheme rules, there may be rather a lot of decision-making, including, but not limited to:
a) DB pension commencement date (e.g. normal retirement date (NRD), earlier than NRD or even later than NRD);
b) whether or not to take a tax free pension commencement lump sum (PCLS) in exchange for some pension;
c) for deferred members of some schemes conversion to DC/SIPP via a cash equivalent transfer value (CETV)
all of which could significantly impact the actual LTA hit associated with a DB pension. Depending on the choices made at a) to c) above the actual LTA hit incurred could range from approximately half the nominal LTA associated with retiring at NRD to about twice that value. So quite a range of values, and, I think, more important if DB benefits make up a relatively large proportion of your total retirement assets.
I agree the DB LTA hit would in principle be the same irrespective of which of your rundown strategies you choose – but the consequential impact to taxes paid, etc may be non trivial.
Thanks for your thoughts.
Those are all fairly simple calculations in comparison to what I was looking at in the post. I’ve written about CETVs before and indeed, I’ve just completed my first transfer.
I’m not a fan of DB pensions because I’m not close to my FIRE number. At the moment you would have to live a very long time to do better from a DB pot. I think if your situation is marginal (or you have a low risk tolerance) the perceived security of DB pensions might be attractive.
Indeed, all discussions of this nature are rather specific to an individual(s) situation including: financial where with all (inc. risk), circumstances, perspectives (including conscious and unconscious biases), etc.
When you say you are “not close to my …” I am assuming you must mean that you are well over rather then under your FIRE number when the FIRE number is expressed as a multiple of your annual expenditure. Is that correct?
Out of interest, did your recent CETV transfer include the possibility of a partial transfer – or was it just take it or leave it?
True, but what I meant was that – for example – a CETV calculation can be done in your head. You have an SWR, which converts to a break-even multiple of the annual payment, and they either offer you more or less than that. Obviously there’s a (personal) grey area around the break-even, but you don’t need a spreadsheet.
Yes, several times my FIRE number. CETV transfer was all or nothing, but on a small pot at a high multiple.
Great article Mike, thank you for sharing your valuable knowledge. I have a sizeable DB due at 60 and a SIPP that I am loading up with the AA each year and may hit LTA, do you or any readers have a view on whether it is best to take DB before SIPP given their generous calculation or better to clear SIPP first if threatening LTA and wishing to balance minimising tax and maximising flexibility?
I think a lot depends on the flexibility and extras in your DB scheme – every one is different.
SIPPs are normally accessible earlier. The key here is to either (1) have a plan for using up the LTA before age 75, or (2) wait for a market pullback and transfer your funds to draw down when they are under the LTA (this of course limits future contributions).
Personally I would be looking to transfer out of the DB pension while the cash values are so high, but everybody’s circumstances are different. If you do that you end up with a bigger SIPP, so I’m not clear on what you mean by “take DB before SIPP”.
Thank you Mike. I can access both at 55 but take
a hit on DB by c5% pa every year I take before 55, though that reduces my LTA and I get the money early. No good offer for DB cash value exchange unfortunately. I was weighing up whether I front run SIPP at 55 and leave DB until 60 or take DB early and take SIPP later.
*that should read c5% before 60
Mike, I am rapidly approaching an LTA decision point — I am 58 and my SIPPs are pretty well right at the LTA just now. Could you maybe comment on an alternative drawdown strategy that I have been mulling.
The core idea, simply put, is that once you pass the LTA, all your pension’s growth flows into LTA penalty zone, but (nearly) every tax rate you might encounter outside a pension will lower than the effective rate inside a pension when you add in the LTA penalty. Even for a consistently higher-rate taxpayer, capital gains tax is 20% but with an annual allowance and deferrable if you don’t sell, dividend tax is 32.5%, and income tax 40%. No blend of these comes to more than 55% (and it’s unlikely you could reach 55% even if stumbling into the appalling effective 60% bubble band).
With that in mind, my own plan is to crystallise my entire pension all in one go right at the LTA. This splits it into two parts, the 25% PCLS and the drawdown portion.
The PCLS part can then be reinvested into the exact same funds as it came from. The dividends it generates will be taxable annually, but at 7.5%/32.5% (lower rate/higher rate) and so much less than 40%/55% LTA penalty. Any recognised capital gains might be taxed at 10%/20% at the worst, where they do not fall into the annual allowance, so again much less than 40%/55% LTA penalty. There is also the option to perhaps slowly funnel this into ISAs for greater tax efficiency.
Any (deferred) withdrawals from the drawdown part are taxable at worst at some blend of 20%/40% income tax rates, but again much less than the 40%/55% LTA penalty. And providing all the nominal growth in this part is drawn at normal income tax rates before age 75, there is no LTA penalty due then either. (Even if you take no drawdown at all until the day before you hit age 75(!), it will almost certainly be worth taking it all then as fully taxable income, since the 20%, 40%, 60%, and 45% blend on it is very likely to undercut the likely 55% that is the combination of 25% LTA penalty plus later income tax you would face after the age 75 hurdle.)
In practice, this route to things will entirely avoid any and every LTA penalty payment whatsoever. For an investor with no other income, the cost is that some drawn income may start to push into higher rate income tax, but it seems at least possible that this avoidance of a nasty LTA penalty at age 75 may be worth that offset.
It also makes efficient use of any fixed protection, because it fully uses the entire larger LTA before inflation has been allowed to ravage it. No protection is ‘abandoned’.
Probably fiddly to model through spreadsheets, but it would be interesting to see how it compares with your approaches above. Anyway, any thoughts? Thanks.
There’s no doubt that moving into drawdown is the way to protect against the LTA. I have too much money in the SIPPs (mine and my partner’s) at the moment, so in a way I’m waiting for a market crash to take advantage. UFPLS is essentially a holding pattern until we get there.
The problems with drawdown are (1) what to do with the cash lump sum and (2) how to coordinate between the two halves of a couple. My other half is still working and so can’t go into drawdown yet – so I have to do UFPLS while we wait. Both cash lump sums will probably end up in a seaside cottage.
The point of this article was not not compare UFPLS with drawdown (as most of the comments have done), but to explore whether there was a “magic bullet” withdrawal pattern that would negate the need for eventual drawdown. Turns out there isn’t.
Understood, Mike. Thanks for the explanation.
As for how to handle the PCLS… I’ll simply put it back into whatever asset classes it came from. The combination of dividend tax plus any capital gains (should be minimal as a passive tracker investor) will certainly be much lower than if still inside the SIPP and so subject to the LTA penalty.
The rest will sit in deferred drawdown until I need it, again all in the same funds. In essence then, the only thing that changes with this chunk is that I have jumped the LTA hurdle at the optimum point and before penalties kick in.
The LTA really is a planning bear, isn’t it?
I’m not sure that the taxes would be minimal enough for me. I like the way that UFPLS lets me park £20K pa in an ISA. And when we buy the cottage we can offset rental income against a loan.
The LTA is stupid. You shouldn’t have contribution limits and withdrawal limits on the same product. And the level is now too low, given current annuity rates at age 55.
Tax planning generally is a pain.
Okay Mike, and thanks for the follow-up.
I think we differ in that due to other unsheltered investments and a likely sizeable income bump from a 401k I have from working in the US (beware RMDs!), I will consistently be in higher rate tax throughout, at least as far as I can guess. So my tax outside the pension would always be less than that inside. Use up all the LTA now seems like the best route for me.
For you, if taking larger distributions pushes you up a tax bracket
on withdrawals, that may argue towards a longer and slower use of the LTA than in my case. I’m just not crossing inflection points. Probably.
Maybe it’s just me, but I’m personally a bit sceptical of the ‘wait for a market crash’ strategy. If it only turns up after a preceding and offsetting market rise you don’t come out ahead. And even if it does come along exactly how and when you need it, you still have to call it right. I’ve long accepted that predicting markets is impossible, so it seems risky to me, but fortunately I don’t think I am in a position to have to consider it anyway.
If I had my time over I would not save hard into pensions again. The LTA reductions are effectively a retroactive tax. I really despise George Osborne for burdening us with all of this crap. If there is a circle of hell reserved specially for idiotic politicians, he deserves immediate admission. 🙁
Hi Mike,
I am wondering if you have had a chance to reflect on your strategy in light of the LTA threshold being frozen until 2026? It seems to me that the likelihood of the LTA being abolished is pretty much nil. If the Conservative government is intent on using the LTA as a tax raising measure, it is highly likely that a government of any other shade will do the same.
I regularly reflect on the strategy as I have to fill in a lot of forms each year to get my hands on the UPFPLS payment. The strategy hasn’t changed in that I still intend to exhaust my LTA before I hit 75. And as long as my other half continues to work, the most flexible way of implementing this is an annual UFPLS payment.
I no longer think that the LTA will be abolished, but Covid has raised bigger tax issues than that. We no longer have a Conservative government (in tax terms) and I am becoming increasingly resigned to emigration at some point.