USS - General discussion

Universidad

d6fs1l said:

And we wonder why people find pensions difficult....

Three years of in depth learning about pensions, and the main thing I've learned is that I know almost nothing about them. 

It's a weird space to be in when I know more about pensions than anyone in my orbit, and less than anyone who actually knows about pensions. 

Barralad77

Universidad said:

d6fs1l said:

And we wonder why people find pensions difficult....

Three years of in depth learning about pensions, and the main thing I've learned is that I know almost nothing about them. 

It's a weird space to be in when I know more about pensions than anyone in my orbit, and less than anyone who actually knows about pensions. 

It’s verging on the perverse at times. I can’t help but think it must be possible to make it simpler and more transparent (which is what I started out believing it was). By way of an example, compare the pricing of train tickets in India vs UK. In India the price of the ticket seems to be based on class of travel and distance (or at least it was last time I checked). Alternatively, here in the UK the pricing structure (such as it is…) is all over the place and the price of a ticket to a nearer destination can be more expensive than one farther away, and depend on which day you want to travel, and which time of day, and - for all I know - whether your name is Derek or not.

I digress….

Regarding the AA in a DB scheme it seems it’s based in the increase in the value of the pension over a 12 month period. Given that the benefit is defined (e.g., 1/75 plus 3/75 as a lump sum) it seems quite obvious how much it’s increased by. And yet that certainty appears misplaced because…. because somehow we’ve managed to make it completely impenetrable for anyone who doesn’t work in this sphere of expertise. 

The use of CPI seems bizarre if the value of your salary - and hence the 1/75 faction of that salary - hasn’t increased by CPI, which in many cases (most?) in USS it hasn’t. If the cap(s) imposed by USS on the amount by which the pension is uplifted each year means that the uplift doesn’t match (it’s less than) the CPI then how is the CPI relevant to calculating how much the value of my pension has increased by?

d6fs1l

Barralad77 said:

The use of CPI seems bizarre if the value of your salary - and hence the 1/75 faction of that salary - hasn’t increased by CPI, which in many cases (most?) in USS it hasn’t. If the cap(s) imposed by USS on the amount by which the pension is uplifted each year means that the uplift doesn’t match (it’s less than) the CPI then how is the CPI relevant to calculating how much the value of my pension has increased by?

The idea of the CPI uplift in opening value is to remove from your utilisation of the AA an amount of benefit which is simply equivalent to the indexation of your previously accrued pension. So it actually "helps" someone who wants to use their AA because it leaves more of it available for further contributions elsewhere. (That's why I'm bothered by the seeming omission of it in the USS figures, when it makes such a difference in the past two years.)

There's a horrible grinding of conceptual gears in this, though -- which is what happens when something like the AA (which is clearly designed for the DC world) is applied to the DB world.

Barralad77

Which poses an interesting question (for me, at least). What would happen if, according to the USS calculation method my AA figure was £70,000 but where using the alternative CPI-indexed calculation method it was, say, £55,000? What would USS do? I guess the first thing USS might do is to see whether I have - according to their records - any unused AA from previous years (in which case it might not be an issue for them) but what if this scenario applied for, say, 4 consecutive years? As far as USS is concerned I’ve exceeded the AA by £10,000 per year but according to the other method I’ve stayed within the AA the whole time. Or would USS simply stop accepting contributions at some point?

I’ve no idea - does anyone?

But I’m still not convinced that the use of CPI is appropriate. When I took the value of my 2024 pension earlier and increased it by 6.7%, then compared it to the value of my 2025 pension, the difference was less then £40. Given that the valuation of my pension across the 12 month period rose by over £900, how does the CPI calculation method make any sense? The increase of over £900 is equivalent to 1/75 of my salary (which does make sense). If it’s correct that the AA is based on the increase in the value of the pension over the PIP/year then in my case the increase is clearly - as far as the DB element goes - ~£900, not <£40.

Framing it another way, what would the AA be if the following applied to someone with a salary of £45k who didn’t contribute to the DC element (the IB in USS):

Opening value £10,000
CPI - 10%
Closing value £10,600

Once the opening value has been increased by CPI it becomes £11,000 and that’s lower than the closing value, meaning you end up with a growth figure of -£400, even though it’s clear that the pension has increased in value by +£600.

What happens here? Does this person still have all £60,000 of their AA available? Not according to the USS calculation method, which seems reasonable.

kermchem

Barralad77 said:

Which poses an interesting question (for me, at least). What would happen if, according to the USS calculation method my AA figure was £70,000 but where using the alternative CPI-indexed calculation method it was, say, £55,000? What would USS do? I guess the first thing USS might do is to see whether I have - according to their records - any unused AA from previous years (in which case it might not be an issue for them) but what if this scenario applied for, say, 4 consecutive years? As far as USS is concerned I’ve exceeded the AA by £10,000 per year but according to the other method I’ve stayed within the AA the whole time. Or would USS simply stop accepting contributions at some point?

I’ve no idea - does anyone?

My understanding is that USS themselves do not care if you exceed the £60K Annual Allowance. They simply do the calculation for each member and report the number to the member and to HMRC. If you exceed £60K in one tax year then you can use unused Annual Allowance from previous years - and this then becomes a problem if you have used all the carry forward that you are allowed.

If I look at my own Annual Allowance used according to USS versus my salary, then a USS member will need to be on a salary well over £100K to be in danger of using £60K in Annual Allowance. At that level I assume that HMRC will have them down for self-assessment, and a possible tax charge if they exceed the Annual Allowance with no carry forward to use. USS will pay the tax charge for you, called "scheme pays". And USS allows high salary members in danger of exceeding AA to use a Voluntary Salary Cap.

My understanding is that when the AA was £40K the headline problem was that most hospital consultants over the age of 55 were exceeding the allowance every year and threatening to take at least partial early retirement. Raising the annual allowance to £60K was yet another action needed to "save the NHS". Not relevant to this thread, but my guess is that if any of the wilder ideas being kite-flown for the forthcoming Budget prove to be even remotely true, then bashing the pensions of the middle classes will see more hospital consultants looking at improving their golf handicap. 

Barralad77

kermchem said:

If I look at my own Annual Allowance used according to USS versus my salary, then a USS member will need to be on a salary well over £100K to be in danger of using £60K in Annual Allowance.

Not so sure about that. Someone on £80k could put £4,000 into the IB each month (48k) and then their salary/75 x 19 (20,266) makes their AA £68,266 using the USS method. I can easily imagine people doing that (I was putting in nearly that before I dropped my hours). Very much a first-world problem, as they say, but not beyond the realms of possibility.

kermchem

Barralad77 said:

Not so sure about that. Someone on £80k could put £4,000 into the IB each month (48k) and then their salary/75 x 19 (20,266) makes their AA £68,266 using the USS method. I can easily imagine people doing that (I was putting in nearly that before I dropped my hours). Very much a first-world problem, as they say, but not beyond the realms of possibility.

Hmm, yes, interesting numbers you picked. I am not quite on £80k, and I am currently shovelling enough into IB each month to get me just under the 40% threshold this tax year as I head for retirement next year. I am not going to reach the AA limit, and have plenty of carry forward if I did. I would hope that anyone doing the same as me, or your £80k example, knows about the AA and what they are doing. In most cases, £48k into IB for more than 3 years would cause other problems - like not able to get the resulting funds back out in a tax-efficient manner. Hence dropped hours?

Barralad77

I think we’re back where we started. Q. Does USS calculate AA the right way or not? If they refer someone to HMRC because they have exceeded the AA according to their method of calculation then unless HMRC check that data then tax will be paid. But if we use the method of calculating AA on the Govt’s own website then they may not have breached the AA of £60k and will have no such tax to pay.

This is the point made by @d6fs1l, earlier. USS appear to be calculating the AA in such a way that means someone reaches the £60k limit sooner than if it were being calculated by the method that is - presumably? - endorsed by HMRC.

Which method is correct? The one where the opening value is increased by CPI then compared to the closing value or the one where you divide someone’s salary by 75 then multiply it by 19 (adding any IB contributions in both cases)? They give startlingly different outcomes and they can’t both be correct (can they?).

Barralad77

kermchem said:

In most cases, £48k into IB for more than 3 years would cause other problems - like not able to get the resulting funds back out in a tax-efficient manner.

It would be tax-efficient one to some degree or another. If you will be a lower rate tax payer when you come to take it out the money in the IB will either be tax-free (if you take up to 3.667 x the value of the annuity at the point of retiring) or you’ll pay a net tax rate of 15% (due to getting a quarter of it tax-free and paying 20% on the rest) as long as you take it in stages. So even if some of the £48k put into the IB came from the 20% income tax band you would still see a gain (moreso given that the IB contribution would also secure NI relief as well as tax relief).

d6fs1l

Barralad77 said:

Which method is correct? The one where the opening value is increased by CPI then compared to the closing value or the one where you divide someone’s salary by 75 then multiply it by 19 (adding any IB contributions in both cases)? They give startlingly different outcomes and they can’t both be correct (can they?).

The relevant legislation for calculating the pension input in respect of a DB scheme is s234-236 of the Finance Act 2004 and at a quick look it supports the more detailed commentary in HMRC's internal manual (as one would expect). The CPI uprating provision is at s235.