Rules on using Occupational Pensions Revaluation Orders

timodell

I have been told by my pension scheme administrator (Willis Towers Watson) that the indexation of the components of my deferred DB pension which are subject to statutory price revaluation works in the following (bizarre) fashion: If, like me, you left employment on 31st August 1990 and want to calculate the indexation factor to potential retirement dates between, say September 2018 and September 2019, then you need to look at the annual "Occupational Pensions (Revaluation) Order" which is published every November for retirement dates in the following calendar year. For retirement dates after the anniversary of the date of leaving service (eg. 1 Sep 2018) you would use the row of the order published the previous year for the period starting 1st January 1990 - in this case the factor in the 2017 table of 121.4%. For dates in the year prior to the anniversary of leaving (eg 1 May 2019) you would use the row starting 1st January 1991 - in this case the factor in the 2018 table of 104.4%. But for retirement in September 2019 I would then go back to the row starting 1st January 1990 - in this case 126.7%. So the rate to be paid drops arbitrarily by around 8% if taken earlier in the year than the anniversary of leaving service.


Effectively, what is happening is the inflation rate for the year in which you left service (in my case 10.9% in 1990) is omitted for retirement dates between 1st January and the anniversary of leaving service, but then is reinstated from that anniversary until the end of the calendar year. This saw-tooth pattern is repeated every year. For anyone with a date of leaving service in a "high" inflation year like 1990, this gives a massive advantage for retirement late in a calendar year.


If true, this quirk of the legislation seems to make the exact date of retirement within each calendar year absolutely crucial to maximising retirement income (as once in place, this "up-down" starting value is then only ever increased by current inflation, so if you miss out the "high" starting inflation rate from your leaving year, you never get it back!).


What is even more annoying, is that WTW will wrongly base any retirement quote they issue late in a calendar year (but for a starting date early the following year) on the factor from the old table - so in my case, a quote for a start date in May 2019 included in a letter dated December 2018 was based on the 121.4% factor, but when refreshed in January 2019 (for the same May 2019 start date) is dropped by 8% because their system only now picked up the 104.4% factor - even though it had already been published in November 2018.


Both the legislation, and the WTW procedures seem mad to me...

zagfles

Yes, it's a know foible in the system. If inflation was high in the year you left and is low now, your pension will drop on 1st Jan and increase on the anniversary of leaving. So you're better off retiring late in the year, between the anniversary of leaving and the end of the year.

There's not really an excuse to get this wrong after the tables are published - before then it's guesswork.

timodell

What an incredibly badly designed system...anyone who left employment in 1988, 1989 or 1990 (when inflation was 5.9%, 7.3% and 10.9% respectively) must be really careful to make sure they put their pension into payment after the anniversary of their leaving date or they will lose out by between 4% and 8% of their annual pension which could cost them a five figure lump-sum equivalent (for me the difference amounts to £750 pa equivalent to a lump-sum hit of c£30k). Conversely, someone who stopped work in 2009 when inflation was -1.4% will be much better off retiring before the anniversary of their leaving date... Not sure if anything can be done about this retrospectively if you were on the wrong side of this anomaly, but I think you could have a good case for suing your IFA if they didn't point it out to you.

Brynsam

timodell wrote: »

What an incredibly badly designed system..... Not sure if anything can be done about this retrospectively if you were on the wrong side of this anomaly, but I think you could have a good case for suing your IFA if they didn't point it out to you.

Hindsight is always a wonderful thing. As for 'suing your IFA' - even if retirees have an IFA, how likely are they to ask for advice in such a niche area? If the IFA doesn't know it's an issue and the client doesn't raise it, there are no grounds for complaint (and complaining is the sensible route, not suing - the former route is well covered by one or other Ombudsman and can be pursued free of charge).

zagfles

timodell wrote: »

What an incredibly badly designed system...anyone who left employment in 1988, 1989 or 2000 (when inflation was 5.9%, 7.3% and 10.9% respectively) must be really careful to make sure they put their pension into payment after the anniversary of their leaving date or they will lose out by between 4% and 8% of their annual pension which could cost them a five figure lump-sum equivalent (for me the difference amounts to £750 pa equivalent to a lump-sum hit of c£30k). Conversely, someone who stopped work in 2009 when inflation was -1.4% will be much better off retiring before the anniversary of their leaving date... Not sure if anything can be done about this retrospectively if you were on the wrong side of this anomaly, but I think you could have a good case for suing your IFA if they didn't point it out to you.

I would wager most IFAs don't know about this. But most people with occupational pensions wouldn't use an IFA anyway...

jamesperrett

There are all kinds of odd foibles in pensions like this. In my case there was a 14 day window in every 3 months where I would get my maximum salary taken into consideration. I had an inkling that something like that would be the case but no-one would tell me exactly how these things were calculated until I'd set my retirement date (which luckily fell into that window).

squirrelpie

If the IFA doesn't know it's an issue and the client doesn't raise it, there are no grounds for complaint

Are you sure about that? Surely one reason that you pay an adviser for advice is that they should be aware of the facts that are relevant in the area they give advice. I'm surprised if there aren't checklists of such things to be checked.

timodell

I agree with Squirrelpie: if you do employ an IFA, I think they should know about these sorts of idiosyncrasies. But I also agree with zagfles that many would not expect to need one for the simple act of bringing a deferred pension into payment. Of course, the administrators (like WTW in this case) are the best placed to know the implications of this timing quirk, but they are in the clear because of their "small print" about not giving advice and not being liable for mistakes. :mad:

timodell

jamesperrett wrote: »

There are all kinds of odd foibles in pensions like this. In my case there was a 14 day window in every 3 months where I would get my maximum salary taken into consideration. I had an inkling that something like that would be the case but no-one would tell me exactly how these things were calculated until I'd set my retirement date (which luckily fell into that window).

Thanks for that. Do you think it would be useful to start a "sticky" listing all of these "foibles" so that they are easily flagged to others...?

FloraandFauna

This has blown my mind somewhat.

I have just been trying to work out how I would apply that to my deferred DB in 2032, which is when I'm aiming to retire. I think it goes something like this:
- I left a DB scheme 15th December 2009
- revaluation rates, depending on whether I start the pension pre or post 15th of December 2032 are (in the current revaluation table):
  - 1st January 2009 - 31st December 20[20]: 23.60%
  - 1st January 2010 - 31st December 20[20]: 25.30%

So just pretend for a moment... that the revaluation table numbers above were actually published in 2031...

If I retired in 2032 after my leaving anniversary, i.e. 16th December 2032, I would use the 1st January 2009 - 31st December 2032 row, of the table published in November 2031. Which would be the 23.60% revaluation.

If I retired before my leaving anniversary, i.e. 16th June 2032, I would use the 1st January 2010 - 31st December 2032 row, of the table published in November 2031. Which would be the 25.30% revaluation.

So I would be better off retiring before the anniversary date? Or is there some other calculation I should be aware of? Would that give me the higher revaluation rate, but applied to 1 years' 'less'?

TVAS

The pension is driven by the date of retire you attain the scheme retirement age. You cannot change the age at which you retire therefore the month in relation to the month they use to revalue the pension cannot be changed unless they decide to change it and that would need to be in the scheme rules.

Or
Your age in months and years if you retire early although you will pay a penalty for taking the benefits early. Depending on the scheme some will revalue your pension to the scheme normal retirement date then discount back by the number of years and months to the age of taking benefits. This is generous. Less generous is when the scheme revalues until your actual age and then applies the penalty for age in years and months benefits is taken early. 
Somebody above mentioned leaving the pension before 1990. Schemes are not obliged to revalue the excess pension (the amount over the GMP) in deferment:
Social Security Act 1991

Leavers on or after 1 January 1991

Revaluation extended to cover the whole of the member's pension, in excess of the GMP.  Annual increase applicable was the increase in the Retail Price Index (RPI), capped at 5% (sometimes known as 5% Limited Price Indexation - LPI).

The minimum is above the scheme can do more if they want to. 
With regard to IFA's they would not be interested in giving you advice on a DB plan unless there is a fee however as I said before you cannot change the month when you attain the scheme normal retirement age. 

If you think the scheme has done it wrong you would complain to the PENSION OMBUDSMAN as they deal with occupational schemes.