NHS pension - not sure whether to take standard or maximum lump sum

LightFlare

Moonwolf said:

Mr_Crabbs said:

All, thanks for your comments

Moonwolf said:

... I'm not sure about recycling but you will have extra spare income.

I'm not a fan of taking the lump sum if you don't have a use for it ...

If you lean towards more income has your wife looked at taking the McCloud years in 2015 ...

I haven't seen a good clear resource on McCloud so am struggling to understand it. I'm not sure, though, that any benefit would offset the loss from what you say, unless I'm misunderstanding.

On McCloud.

At a basic level, McCloud lets you choose whether to take the 7 years of service from 2015 to 2022 in the 1995 scheme or the 2015 scheme.

In either case, the 1995 will use your best of last 3 years salary at the time you take it and you will get 1/80th of that for each year you worked.

Someone earning top of Band 4 £29,114 now who had been working at Band 4 for the NHS since April 2005 would either get 10/80ths x £29,114 or 17/80s x £29,114 plus 3 x that as lump sum.  So pensions of either £3,539.25 or £6,186.73 and a lump sum of £10,917.75 or £18.560.18.

However the 2015 scheme accrues by 1/54th and in those 7 years has had revaluations above inflation growing by more than salaries have grown.  In the example above I think the 2015 pension would be £1,679.79 with McCloud in 1995 or £5,860.36 with the McCloud in 2015, plus if the employee was going to work another 7 years then the extra 1.5% growth would accrue an extra £536.92 of annual pension over inflation.

I worked out that, ignoring inflation and assuming pay now follows inflation, and this Band 4 worked to 67, the difference, summing the 1995 and 2015 pensions, from 67 would be

With McCloud years in 1995 pension = £12,040.44
With McCloud years in 2015 pension = £14,000.45

So £1,960 extra a year to offset against £17,832.33 extra pension paid over the 7 years and £7,642 larger standard lump sum, I think that is equivalent to a commutation rate of about 13?

These figures will be totally different for someone who has had promotions or climbed bands in the period and I have thrown these calculations together so there will probably be errors.  However, I think there is a group of people who will be better off taking the years in 2015, mostly people who have not had promotions, many of whom are just looking at the bigger lump sum.

All I'm saying is check the figures carefully and think about what they mean.

I would add - you MUST also factor in the reduction in pension if you are taking it earlier than the age specified for each scheme.

1995 pays full at 60 (some professions may still be 55)

2015 pays full at 67

both work on roughly a 3-4% reduction per year early that the pension is taken - hence why I am glad to be able to have 2015-2023 in my 1995 when retiring at 56. It means my amount is reduced by ~14%

Had those 7 years been in the 2015 section, then the reduction would be in the range of 40%

Due to the accrual rate difference 1/80 and 1/54 - there will be a crossover point. Mine was 59 years old, but everyone’s will be different

If the OPs wife is planning on working untill 67 it may be more beneficial to leave those 7 years in the 2015 section

Het employer should have some form of pensions person that can supply all the info so you can make a more informed choice (you only get one chance after all)

Pat38493

Mr_Crabbs said:

I'm not clear what you mean by 'sequence of returns risk'?

Sequence of Return Risk (sometimes shortened to SORR) is the risk that your investments will be eroded too quickly due to getting negative or low returns, early in your investment period (i.e. retirement).

For example over a 10 years period, if you have £100K invested, and you heard that the average returns are 7%, you decided to take out £10K per year, thinking that due to the 7% returns you will have some left at the end.

Stock markets do not have positive returns every year - some years (roughly one in 4) will be negative.

If your negative years comes in the first years of your plan, you will run out of money because you are taking money out where your investments are in a "distressed" state.  This is basically SORR.

This is the reason why the "safe withdrawal rate" over 30 years from an investment pot is generally considered as less than 4%, even though average stock market returns over the long term are more like 6 or 7% above inflation.  The average doesn't matter if you get all the bad ones at the start.

hugheskevi

Mr_Crabbs said:

hugheskevi said:

It is interesting to consider the breakeven points for a basic rate and additional rate taxpayer in retirement, net of tax ... for someone who is going to be a basic rate taxpayer in retirement, taking the taxed pension is financially better ... for a higher rate taxpayer, the lump sum starts to be attractive. 

Sequence of returns risk will be an important consideration, which this example assumes away as an issue but needs to be acknowledged. A lump sum may also be difficult to invest within ISA limits, this is likely to be much less of an issue with pension income.

Ideally flexibility would be provided for by separate DC pension which is perfect for such a purpose (especially if part of the same scheme as the DB pension, as in the LGPS, and so can be taken largely or entirely tax-free)

You would want to expand this to cover withdrawals from the lump sum and pension income based on personal preferences to see how that affected the breakeven points on different rates of return.

Thanks. Do you have a reference for this, or the source calculations?

Just playing around with my own spreadsheets using the assumptions set out.

I'm not clear what you mean by 'sequence of returns risk'?

Market returns can be very volatile, with double-digit gains or losses a frequent occurrence. A big loss when investments are at their largest amount can be very damaging. So a 5-year sequence of returns that is, say, +2%, +3%, +3%, +2%, -10% might give much better outcome than -10%, +2%, +2%, +3%, +3% depending on how the investment was being drawn down. There are many sources of information and mitigations available online.

DB pension is very safe and very inflexible - a guaranteed inflation protected amount each year. These are valuable features that are complemented perfectly by DC pension which is the exact opposite. DB gives safe, boring, inflexible income whilst DC gives flexibility.

That flexibility is great for tax-efficiently withdrawing large amounts to give access to more funds early in retirement, perhaps to fill a gap prior to State Pension becoming due, or just to have more whilst being more active early in retirement. The same flexibility can be achieved by DB pension commutation, but at a high price. So saving in a DC pension to avoid having to ravage the value of the DB pension through commutation is likely to be a better solution.

I'm not clear what you mean by "can be taken largely or entirely tax free"? We are considering a separate DC (AVCs / additional pension or SIPP) to mitigate the higher rate tax my partner would otherwise pay.

Some DB pensions also have a DC element, and the tax-free amount available across the value of the DB and DC pension can be taken entirely from the DC section (as they are both part of the same single pension). The LGPS is the most common example of this.

Higher rate tax into a pension is usually very attractive, especially if going to be paying basic rate tax in retirement when the lower tax rate combined with tax free lump sum becomes very significant.

I think we need to do some calculations but are unsure how to do it and don't have access to an FA, particularly in the timescale my partner wants to work on! Have you any sources?

I'd just do everything in spreadsheet modelling, and test variability with tools such as cFiresim.

jimi_man

phynix_uk said:

Dazed_and_C0nfused said:

phynix_uk said:

Whilst I agree with the 12-1 commutation not being attractive compared to some, possibly most DB schemes, the instant conclusion that this isn't in your best interest is simply false. Stating that it'll only take you 12 years to essentially break even is also incorrect. The calculation needs to take tax into account and if your projecting forward in time, you need to also include potential growth of the £48,000. If you want to stress test numbers, you need to be fair to them.

Ignoring higher rate tax, and tax mitigation (which would absolutely be sensible) the income received under the standard terms would be taxed so the 12-1 ratio would as a minimum need to be 15-1, pushing the crossover age to 75... then you could argue that the additional £48,000 also grows over time, and if the FA you dealing with manages money well then this would push that age put even further. 

Generally fair points.

But the pension given up is almost certainly only £4k in year one, you can't ignore the annual inflation proofing, whatever that is for this scheme.

Inflation proofing is really the only advantage. 

I wouldn’t say this is necessarily correct. Financial Security of a Govt backed pension is a huge advantage to my mind. 

Mr_Crabbs

Dazed_and_C0nfused said:

I think the potential to take more than 25% tax free from a separate DC pension is something that is specific to LGPS ... But higher rate tax relief and 25% TFLS is still quite an attractive proposition 🙂

I haven't seen any reference re. the NHS pension; and yes, there seem to be distinct advantages

Mr_Crabbs

Hi all, thanks again for your thoughts.

@LightFlare thanks re. taking 2015 early. I'm still struggling to understand McCloud. Partner had to make a choice some years ago and I'm not sure whether that was to stay in 1995 or move to 2015; I suspect it was as she has very little in the 2015 scheme (c. £2k p.a. value). I can't work out whether she will get a McCloud election, whether it will happen after she retires, and whether it will be paid now or after 67 if she takes it; now but with reduction, presumably.

@Pat38493 @h@hugheskevi thanks re SORR etc., I understand, I think. I'll look up mitigations.

The DC flexibility thing is interesting. One of the factors I'm thinking about is that partner plans to retire fully at 65, not 67, leaving a gap of two years before the 2015 pension comes in and then a relatively small 2015 pension, which might be one reason to stick with the standard lump sum and better 1995 scheme payments. She could maybe start to take DC scheme then if she paid in for a few years. It wouldn't be much, but it would be something. Will definitely be lower rate in full retirement.

I'd like to model options but I'm not confident enough about my maths or understanding to be sure I'm modelling correctly. I did try and it looked as though break even was 16 years at 20% tax rate assuming 4% interest. Thanks for the cFiresim limk.

Yorkie1

The McCloud election is made at the point you retire (or take the first part of your pension, if it's in a couple of parts which can be taken at separate times). 

Annual pension statements should show that year's figures for each version of McCloud. And you should get both sets of figures before you choose to retire, to help.

Mr_Crabbs

Morning all,

I think we have white smoke.

I drove myself mad yesterday running spreadsheets with little confidence that I was getting things right. It felt as though you can almost prove anything with numbers depending on the assumptions! However using 'standard' assumptions and given my partner's situation we reached a number of conclusions:

We'd probably want low risk in retirement so wouldn't go for high returns on savings or investments

We could take the £48k and save / invest it at low to medium risk (say 3-5% annualised). If we do that but don't draw it down we don't benefit from it, except as a hedge against late life care costs. If we do it and draw down the equivalent of the commuted pension income, taxed at basic rate, the capital + returns run out in around 14 - 17 years (this seems about consistent with @hugheskevi analysis?)

To make it last, or to get the equivalent of the long term pension income, the draw down would have to be substantially lower or the investment riskier (7.5%+ annualised).

We might be leaving money on the table by not taking the lump sum and investing aggressively but that's not what retirement's about and we are accepting that. If we don't use all of the pension income, it could also be saved / invested to provide an additional return.

While we might want to use the capital to contribute towards a house move, we have other savings and there isn't an obvious, compelling need to take it. There's a slight risk of falling short of a dream house, but so be it. The additional lump sum is only one year's wages for me or my partner, and > 10% of our other savings and assets, so it doesn't make a major difference to our finances unless we are stretching ourselves to buy a house.

We can mitigate higher rate tax in either case (standard or maximum lump sum) using additional / AVC / DC pension via NHSBSA, or SIPP without MPAA or recycling issues, so pension income should be at basic rate.

As far as I can work out, partner's 2015 scheme and prospective additional contributions will be below the tax free threshold for contributions (which I think will be equivalent to her new wage of c. £42k). Any lump sum she would take is not taxable, being well under the £268k limit.

McCloud decision is on or after retirement, and we're still trying to understand it but I'm not sure it would change the lump sum decision now.

Partner prefers the security of a regular, index linked government backed income rather than what she sees as the greater uncertainty and worry of having to save / invest and manage the lump sum.

Partner does not lead a healthy lifestyle, but does not have known health conditions and feels she should survive long enough to benefit from the long term pension income.

So the conclusion is to take the standard lump sum and higher regular pension income from the 1995 scheme, partial retirement at 20% reduced hours, with income reduced below the higher rate threshold using additional pension payments. She's going to take full retirement at 65 and the higher payments will also help bridge the gap to 67.

One thing that did occur to me; she would only have her additional pension for 3 - 4 years. Is there a minimum time limit for pensions?

I've learned a lot from this discussion. I understand things through discussion and kicking thoughts around and it has been incredibly useful for me, and I hope for anyone else wanting information. Thanks @Moonwolf, @LightFlare, @Dazed_and_C0nfused , @Pat38493, @Lowtrawler, @crv1963, @hugheskevi, @Yorkie1 and anyone I've missed.

Mr_Crabbs

crv1963 said:

I do pay 40% tax on total income wage + pension, so initially I paid into SIPP to reduce the HR tax and when the option to rejoin the NHS Pension - 2015 scheme I did so

Lowtrawler said:

There were effectively 2 choices available:
1. Pay into a SIPP
2. Subscribe to Additional Pension (pretty much the same as the AVC you mentioned but retaining Defined Benefits)

I assume both of you / your wife are only doing this for a short time and there is no problem with doing so? It's likely my partner would only need to mitigate higher rate tax for 3  - 4 years. Not sure if there is a minimum time to hold / contribute to a pension, mand what investment if it would only be held for a short time.

crv1963

I plan buying additional pension for 6 years only. You can buy additional pension in £250 slices so can mitigate for 3-4 years, just play with the calculators and your prospective amount of HR tax.