UFPLS vs flexi-access drawdown - pros and cons

RL11

gm0 

EdSwippet

gm0 said:

...
Or if you had slammed the crystallisation in up front with FAD at 55 when it was only £1m before it grows.  Same draw, same value at the end (as same return and inflation/indexation and draw). 

Your projection might be a bit off, based on this assumption. You've assumed (I think) that the 25% PCLS taken at age 55 is invested in the same stuff outside the pension as inside. Fine, however if not in an ISA -- and the sum involved would be much too large to be pushed into an ISA except over many years -- the returns on these investments will (depending on your other investments and general tax situation) be subject to annual tax on dividends, and also perhaps eventual capital gains tax. Your example may have missed this added tax drag.

However, accounting for this won't change the result, just narrow the differential. The core point is that the LTA penalty creates a situation where tax on gains that accrue inside a pension that is above the LTA is higher than tax on gains that accrue outside. It directly follows from this that fully crystallising and moving the entire tax-free PCLS out when a pension reaches the LTA, and then drawing enough from the drawdown element at normal tax rates to avoid the spiteful LTA penalty at age 75, is the optimal drawdown course.

So yes, for pensions at risk from the LTA (and where drawdown rather than IHT bypass is the aim), UFPLS is an inappropriate drawdown strategy.

zagfles

gm0 said:

For completeness - the interaction with LTA has come up on prior threads on the different methods and it does skew the findings for the small group of people with LTA issues already appearing at pension commencement. Very broadly - people with 75-80%+ of LTA in their 50s.....The issue on choice of mechanics being around managing the 25% additional tax at BOTH the point of crystallisation (via % LTA consumed at each event) and again at 75 for the cash value of crystallised benefits and any remaining uncrystallised benefits.

Doing FAD first "up front" - compare crystallised amount to current personal LTA to determine % used.  For 1m = 100%
Crystallised fund growth thereafter 55-75 only needs to stay at a cash value below the indexed LTA at age 75 (the second test) Ensure you draw enough and pay your income tax - basic or higher rate as usual and all is well.  
So for pots around LTA "marking a lot for drawdown" via FAD becomes materially superior to UFPLS to remove future growth + inflation (in protection scenarios) from the first crystallisation BCE tests.  Drawing enough income to manage fund cash level deals with the 2nd test.  No uncrystallised funds at 75 and no excess funds at 55.  Limiting case at LTA £1m as designed.  No penalty at the limit.  Seems logical.

With UFPLS - if large portions of a pot are already growing around and above LTA AND are left uncrystallised well into early retirement (if for example a first 5% nibble is taken with UFPLS) the growth of the 95% (uncrystallised for 1-2-3-n years) is now hit with the BCE and LTA tests *after* it has grown i.e. when it is crystallised in a much later year.  And then the cash value is tested at 75 by the same rules either way AND the crystallisation test occurs on the portion not yet crystallised less any LTA % not yet used.  So clearly there is a chunk of growth that gets hit for the additional LTA 25% tax at age 75 with UFPLS which wouldn't be hit if FAD was to crystallise the benefit before the growth occurred. 

So if this is true (key point is treatment of uncrystallised benefits at 75 - I am a consumer not a professional reader of HMRC runes......consider the following.  I would not like to be the IFA who told the individual with an LTA level pot to use UFPLS to pull 12.5/37.5=50k each year crystallising 5% of 1m for 20 years and then having a 170k tax bill at the end on the residual fund which then gets crystallised at 75 with no LTA left.  Where did these numbers come from.....

A quick spreadsheet for age 55-75 - crystallisation timing, income drawn and returns and annual fund values and taxes paid (BCE events, Self Assessment) with inflation visible so it's all cash values - brings this to life for any starting fund value and growth and inflation and income assumptions.   The extra penalty tax for UFPLS turns up at 75 BCE when you have run out of % LTA but not out of uncrystallised pot.  You care about this or you don't depending upon your view of death and taxes.

Mechanics - for one (or more) crystallisation BCE a % of LTA is consumed by an amount £x each time you do it until 100% is reached and the age 75 test is then about cash (if you are all done by then) but it is also a backstop crystallisation test to stop it the pension being a handy IHT tax shelter for the very wealthy to load up with uncrystallised benefits outside the estate rather than a pension to actually use in retirement.  

So to my worked example. With £1m at 55 and UFPLS you have 2% inflation and 3% return and the government keep CPI at 2% for indexation (no games) and the LTA indexes with the draw.  Then you draw an indexed 50k and you have circa 707k fund value age 75 which is still uncrystallised in this scenario and scraps of LTA left (rounding) - based on the UFPLS draws having drawn circa 5% LTA each time.  Unhappily there is circa £170,000 penalty tax to pay on the not yet crystallised funds of 707k  which are all above LTA because it has all been used in the 20 slices.

Or if you had slammed the crystallisation in up front with FAD at 55 when it was only £1m before it grows.  Same draw, same value at the end (as same return and inflation/indexation and draw).  But nothing to pay out of the 707k - cash value well below the LTA for the crystallised benefits value comparison and nothing left to crystallise as it was all done 20 years prior.

With these assumptions the initial starting pot where this "extra tax" arithmetic disappears is about 75% or 750k.  I am  not for one moment suggesting this is a sensible drawdown plan in either mechanic. It is just arithmetic applied to the tests and rules as I understand them and compounding doing its work.

Now somebody may well be along in a minute to tell me I have got the rules wrong about the 75 test and crystallisation of the residual uncrystallised benefits.  I do hope so - it seems monumentally arcane and unfair if my understanding is *correct* that the method difference and compounding could generate such a different result for the same starting pot, return and inflation for FAD and UFPLS.  If I am wrong I would like to understand why.

If I have made an error it will be on the treatment of the uncrystallised excess at age 75 but I think from reading that the backstop test of uncrystallised benefits is a real thing for the reason I hinted at above and it would catch you out if you did this the "wrong way". 

Clearly one should look at this with assumptions to taste on income, return, inflation and LTA indexation (should be the same but 20 years is a long time and remember RPI and CPI ongoing fuss).  There are two things being optimised - amount crystallised when and cash value at 75.

OP may not be above 75% LTA in the years approaching commencment but others may be and this particular area pays for study or quality IFA advice if one is lucky enough to be "in the zone".

Again, this seems to be over analysing the situation! Clearly if you're at the LTA, and you expect growth to exceed indexation of the LTA, you're likely to be better off crystallising fully so the LTA is assessed before further growth. Taxes on unwrapped growth are likely to be less than LTA on growth, and avoiding the second test at 75 should be possible with sensible drawdown.

The situation is different if you're at 75/80% of the LTA, there you can just wait and see what happens. No problem taking UFPLSs and keeping an eye on things, then if/when getting close to 100% consider full crystallisation. It's not like once you start taking UFPLSs you're committed to continue.

It's also different if you're at say 110% LTA, then depending on objectives etc you might decide to delay crystallisation as a hedge against market falls - fund goes down and the hit isn't as bad as the LTA charge reduces, fund goes up, more LTA tax but only because you have more in the fund - so good news overall.