Noobie question on required fund size and SWR

For investors outside the US. Personal investments, personal finance, investing news and theory.
Sister forums: Canada, Spain (en español)
---------------
Post Reply
Topic Author
monsteruk
Posts: 2
Joined: Mon May 13, 2024 2:08 am

Noobie question on required fund size and SWR

Post by monsteruk »

HI everyone,

I am new to the forum. Had a quick search and could not find an answer to my question so thought would ask it now (apologies if it has already been answered previously).

So, one thing I am struggling with related to working out the pension pot size I need to fund my retirement.
I am looking to retire early (56) and planning using a 40 year time horizon for me and the wife. I know my income needs annually (these differ at different stages of retirement), so I am trying to work out (based on a conservative SWR) how large a pension pot I need to fund my retirement. The issue I have is that my income needs are not the same for every stage of my retirement. In the early years (kids still about and no state pensions) I need more - lets say £50,000 pa. At the next stage of retirement (kids gone) I will need lets say £42,000, one state pension kicks in and I only need £31,000 and so on down to £20,000 or so for the final 15 years of the horizon.
If I try and plan based on the initial £50,000 pa (the starting annual need) as the basis for my calc and use a conservative SWR (2.55%) it determines I need an enormous pot based on it thinking I need £50,000 a year for the entire 40 years (which I do not). The calc I am using is: 1 / SWR x 100 = Annual Income Multiplier

So my question is... how do I tackle the above issue of different annual income needs at different stages of retirement? I have worked out (in today’s terms) the average annual income need over the entire time horizon (annual need each year in all 40 years divided by 40) which comes up with an average annual need of £30,000 (in todays terms). Do I just use that average as the basis of my SWR / pot size calc (bear in mind that in no one year do I ever take out "the average")? Or is there a better way to do it?

Thanks in advance everyone.
Valuethinker
Posts: 49350
Joined: Fri May 11, 2007 11:07 am

Re: Noobie question on required fund size and SWR

Post by Valuethinker »

monsteruk wrote: Mon May 13, 2024 2:28 am HI everyone,

I am new to the forum. Had a quick search and could not find an answer to my question so thought would ask it now (apologies if it has already been answered previously).

So, one thing I am struggling with related to working out the pension pot size I need to fund my retirement.
I am looking to retire early (56) and planning using a 40 year time horizon for me and the wife. I know my income needs annually (these differ at different stages of retirement), so I am trying to work out (based on a conservative SWR) how large a pension pot I need to fund my retirement. The issue I have is that my income needs are not the same for every stage of my retirement. In the early years (kids still about and no state pensions) I need more - lets say £50,000 pa. At the next stage of retirement (kids gone) I will need lets say £42,000, one state pension kicks in and I only need £31,000 and so on down to £20,000 or so for the final 15 years of the horizon.
If I try and plan based on the initial £50,000 pa (the starting annual need) as the basis for my calc and use a conservative SWR (2.55%) it determines I need an enormous pot based on it thinking I need £50,000 a year for the entire 40 years (which I do not). The calc I am using is: 1 / SWR x 100 = Annual Income Multiplier

So my question is... how do I tackle the above issue of different annual income needs at different stages of retirement? I have worked out (in today’s terms) the average annual income need over the entire time horizon (annual need each year in all 40 years divided by 40) which comes up with an average annual need of £30,000 (in todays terms). Do I just use that average as the basis of my SWR / pot size calc (bear in mind that in no one year do I ever take out "the average")? Or is there a better way to do it?

Thanks in advance everyone.
All you can do is actually put all of this in a spreadsheet -- map out the 40 years, year by year.

Apply an inflation factor. Say 3% pa?

See what you can do.

The problem lies in returns. Whereas we can probably give you pretty reasonable estimates of equity and bond returns over the next 30 years say (equities 3-5% real, bonds c 1-2% real), we don't know the order of those returns year-by-year. Some years will be truly dreadful (2000-03 or 2008-9 or 2022) and some years will be fabulous. Indeed the UK stock market dropped c 82% in real terms during the early-mid 70s bear market.

Usually the way to tackle that is Monte Carlo analysis. But it's dependent on whatever you feed into it - so if your data is skewed, so too will be its distribution of outcomes. I believe people here are able to do MC on their US retirement positions.

Some financial advisers may do MC analysis for their clients.

An alternative, is to be really dead safe with your savings. One possibility is 100% Index Linked Gilts - directly held, not funds, because of the tax position (no capital gains on inflation appreciation). You then *know* pretty much what your Safe Withdrawal Rate will be. Unless the tax regime were to be changed, of course. I find that hard to stomach - it's placing all bets on the UK government staying solvent for 40+ years. After some of the recent gyrations and policy errors, that just doesn't seem to me to be 100% certain any more. But you could do something like 50% ILGs and 50% global equities. Your main issue would be creating that ILG ladder out 40-45 years.

I would say 2.5% seems pretty reasonable. Given say ILGs paying c 2% pa right now? Feels like this is doable. You can price that against RPI-linked annuities to get a sense of cost. However I am not sure an insurer will quote you for under age 60? But it's worth seeing. Again, that will give you a pretty good idea of the SWR.

Don't forget you will probably want to make capital transfers to your kids? At least if you ever want them to own their own homes, you will. I don't see how, for most of today's youth, they will ever own their own homes without parental support for the downpayment. Whether it's £50k in Newcastle, or £120k in London, that first 20% is the hard 20%. Unless we decided to develop all the urban golf courses, dig into the Green Belt, etc-- about as likely in British politics as electing Jacques Delors as President-for-Life of Great Britain, in my view, with the support of the Sun newspaper.
Stork
Posts: 179
Joined: Wed Feb 17, 2021 9:44 am
Location: Portugal (EU)

Re: Noobie question on required fund size and SWR

Post by Stork »

Welcome here.

The subject has been treated extensively by www.earlyretirementnow.com where you can find a google sheet and plug in your own numbers. Do notice that the perspective is US based, but a lot of the basics apply.

Closer to (your) home, the site www.monevator.com also had an analysis of the issue.

In short, with a portfolio of 60% World equity and 40% local or hedged bonds, about 3.25% should be very safe, also for 40 years - if history is anything to go by.

You do have to be aware that the US life expectancy is lower than most places in Europe.
tgp
Posts: 4
Joined: Tue Apr 02, 2024 1:42 pm

Re: Noobie question on required fund size and SWR

Post by tgp »

Hi - see my post from earlier today (Retirement Calculators for UK based Investors). I have built a tool to help UK investors with this - the link is on that post.
Last edited by tgp on Mon May 13, 2024 10:22 am, edited 1 time in total.
User avatar
Svensk Anga
Posts: 1689
Joined: Sun Dec 23, 2012 4:16 pm

Re: Noobie question on required fund size and SWR

Post by Svensk Anga »

One way to size this is to start with your long term draw of 20k and divide by your SWR of 2.55%. This is your long term portfolio and runs at your selected asset allocation. Add a pot of funds to cover the 30k additional you will need times X years for the first phase when you need 50k total, another pot for phase 2, a third for phase 3. As a first pass, consider investing these pots conservatively, especially the first as you cannot count on stock returns in the short term. You could be very exposed to sequence of returns risk in the early years when total portfolio draw is high. The conservatively invested pots get spent down but the long term portfolio gets rebalanced to your target allocation. This is called a bond tent or a rising equity glide path.

You could try to shrink the size of those early supplemental pots of money by allocating toward higher return investments, but then you are taking risk that a bad sequence could spoil your retirement. Maybe phase 3 is far enough out that it could be invested at your long term allocation.
Whether that is acceptable risk depends on how much discretionary spending you could cut, whether you could go back to work, cut baseline expenses, etc.
Topic Author
monsteruk
Posts: 2
Joined: Mon May 13, 2024 2:08 am

Re: Noobie question on required fund size and SWR

Post by monsteruk »

Hi everyone, and many thanks for your responses. I am going through the various information provided to ensure I properly understand it. One conclusion I have already reached (which answers my initial question), is that it is not safe to use an "average" of your annual needs over the 40 year period in place of using the specific varying annual values (although they both add up to the same total fund pot). Obviously, the order in which you draw out the money will impact the pot's ability to survive the course - with front loaded higher withdrawals being more likely to fail than a simple "average".

I used the calculator provided in a response above: https://www.retiresmartcalc.com/#/calc which I found really useful. To test my theory I ran two simulations using the calculator (both same params - 60% stocks and 40% bonds). One scenario where I took my 40 year average amount from the pot each year for 40 years (so the same amount each year - the averaged out scenario). The UK Global back testing showed a 3.9% failure rate for this average approach. However, when I enter the same pot size but this time I enter the actual percentages i need each year (which is front loaded), the UK Global back testing showed a 10.6% failure rate. So the actual front loaded withdrawal approach seems to be 2.5 times more likely to fail than the linear average approach (even though in theory the amount of money involved is identical in both cases). I guess this is to do with a combination of the sequence of returns risk and more money coming out of the pot at the start of the 40 years (so less to grow in the later years).

All quite worrying in a way though… Taking what I thought was a really conservative 2.55% SWR, it seems there is still an almost 11% chance the plan would fail based on back testing! Wow (good luck to people on 4% SWR)! I do realise however, by implementing things like guardrails, you can be a little more aggressive and adjust as the future unfolds!

Thanks again everyone for taking the time to respond! :happy
tgp
Posts: 4
Joined: Tue Apr 02, 2024 1:42 pm

Re: Noobie question on required fund size and SWR

Post by tgp »

Agree, its because your withdrawal schedule is front loaded that you're exposed to more equity sequence risk than if your withdrawal schedule was constant. You could try the simulation with a higher index linked bond allocation (e.g. 60-70% index bonds, 30-40% equity).
Post Reply