Withdrawal strategy comparisons (adapting US to UK)

For investors outside the US. Personal investments, personal finance, investing news and theory.
Sister forums: Canada, Spain (en español)
---------------
Post Reply
Topic Author
maloflora
Posts: 29
Joined: Sat Oct 10, 2009 4:27 am

Withdrawal strategy comparisons (adapting US to UK)

Post by maloflora »

[Edited to update Amortized Spending Calculator output for change to formula]

Hi all,

I’m nearing FIRE and I was hoping to be able to tap into the wonderful collected wisdom of this board around comparing withdrawal strategies. I’ve been experimenting with a few different approaches and lots of different spreadsheets and want to check the results are all making sense.

A bit of background first:
  • I live in the UK, so I’ve basically filled in every spreadsheet (which are all denominated in dollars) with the equivalent in pounds. Of course, SWRs and return projections will be different in the UK, but only 20% of my portfolio is UK equities
    I’m single, aged 49, and pretty flexible around spending. I suppose an absolute minimum spend a year would be around £40,000 but I’d like to be more extravagant than that, particularly in the early years, after nearly 30 years in a pretty stressful and full-on career
In all the financial scenarios I’ve run, I’ve kept the same broad financial parameters, even though each approach takes a different view of future returns, discount rates and so on:
  • I’m looking at retirement at age 50 with a 40-year timeframe
  • My starting portfolio is £1,500,000 (a mix of tax-protected and taxable but I’ve ignored that for now as I have all taxation questions here)
  • The portfolio consists of 45% global equities, 20% UK equities, 20% UK gilts (government bonds), 10% gold and 5% cash
  • I have now, and intend to keep, a rental property that brings in £650 monthly, or £7,800 annually gross. This includes an allowance for costs, empty months etc
  • I have an index-linked workplace defined benefit scheme that should provide £3,100 monthly, or £37,200 annually from age 60
  • I should be eligible for the full UK state pension which will be £750 monthly, or £9,000 annually at age 67
Because the pensions luckily cover nearly all my essentials from about age 60, I’ve been most attracted to VPW / ABW amortisation approaches. These are brilliant in accounting for future income flows coming in at different times and boosting early spend. So I’ve explored seven different models, trying to keep the assumptions as close as possible and looking at what makes the outcomes different.

I should say up front that I’m enormously grateful to longinvest, siamond, Ben Mathew, Big ERN, Ken Steiner and others for all their work and explanations. I’ve devoured all the ABW, VPW, VPW forward tests, Aloha Joe’s blogs and much more quite avidly to try and understand how PMT and NPV calculations work and so on. I wouldn’t have had a clue otherwise where to start.

So to the results, where I’m focusing on what the first year total spending looks like - i.e. portfolio withdrawal plus £7,800 rental income. I’ve not posted links to the original spreadsheets themselves but I can if that’s helpful. Hopefully, experts here will know which ones I’m referring to:
  • Longinvest’s amazingly simple and easy to use VPW spreadsheet: £102,606 total spend in year 1
  • Big ERN’s actuarial calculator: £91,156
  • Siamond’s Amortized Spending Budget calculator: £97,505. Setting the soft spending gate at 50% brings this down to £99,076. I’ve assumed a 3.1% amortization rate, based on 4% stocks return and 1% bonds return and a 70:30 asset allocation
  • Ken Steiner’s Actuarial Budget Calculator: £84,836, assuming a 4% expected rate of return and 2% inflation
  • Ben Mathew’s Total Portfolio Allocation and Withdrawal Planner with Monte Carlo Simulation calculator: £105,000 (assuming 0% growth in withdrawals). My editing skills weren’t up to changing the rows, and so this is the only simulation that runs on a 45 year time frame: age 50-95
  • Ben’s more simple Total Portfolio Allocation and Withdrawal Calculator: £102,510. For both 5 and 6 I found it slightly trickier to incorporate the other income streams to properly assess total portfolio asset allocation and expected returns as I had to work out how to treat pensions, gold, rental income etc. I’ve used 36% stocks, 6% risky bonds and 59% safe bonds once I’d done some calculations. Using the same 4% and 1% return assumptions I get to an expected real return for the whole portfolio of 2.07%, with the present value of the future income streams of £1,329,756 being added to the (risky) portfolio value of £1,500,000
  • Big ERN’s SWR calculator. This one’s a bit different, because even though it incorporates all the future income streams, it outputs a safe withdrawal rate rather than a year 1 spend to be assessed every year. Setting, slightly arbitrarily, a 5% failure rate, I get a 5.25% safe withdrawal rate leading to a first year total spend of £86,550.
So to some questions:
  • Do people, and particularly the esteemed authors of the calculators, think these results are plausible?
  • Is my approach the right one, particularly given my location and financial situation?
  • Is it clear why some of the results are different? Most are very close, of course, hovering around the £100k mark. But both of ERN’s models and Ken Steiner’s model are significantly lower, perhaps due to different return assumptions?
I should say that I am by nature a cautious soul, so it is highly unlikely I will spend to the limit of any calculator but after a bad day at work I like to speculate on what might be possible!

Many thanks for any help!
Last edited by maloflora on Wed Oct 27, 2021 2:55 pm, edited 1 time in total.
User avatar
Tyler9000
Posts: 740
Joined: Fri Aug 21, 2015 11:57 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by Tyler9000 »

maloflora wrote: Sat Oct 23, 2021 10:50 am I live in the UK, so I’ve basically filled in every spreadsheet (which are all denominated in dollars) with the equivalent in pounds. Of course, SWRs and return projections will be different in the UK, but only 20% of my portfolio is UK equities
...
The portfolio consists of 45% global equities, 20% UK equities, 20% UK gilts (government bonds), 10% gold and 5% cash
...
Do people, and particularly the esteemed authors of the calculators, think these results are plausible?
Welcome! You've clearly thought a lot about this. Nice work on all of the research.

Rather than getting too in the weeds comparing numbers (every tool likely has different datasets, assumptions, and methodologies), the most important detail to me is that you live in the UK. That's going to profoundly affect your withdrawal rates in two important ways that the tools on your list most likely don't naturally account for -- local inflation and exchange rates. It's a complicated topic to describe succinctly, so here's an article that explains these issues in detail: Your Home Country Is Inseparable From Your Withdrawal Rate.

But that's not to say the problem isn't solvable. It sounds like you've already adapted some of the numbers to UK data, which is exactly the right approach. As long as you also cover exchange rates (you need the returns percentages denominated in GBP, not just your current account values) and local UK inflation then you should be in the right ballpark.

And if you don't mind adding one more resource to your research list, I have some tools of my own that I designed specifically to calculate the withdrawal rates for any portfolio in a variety of different countries. Here are two links that pre-fill your stated asset allocation in the UK (just copy and paste the shortcode into the black box, and click "model a different portfolio" if you want to tinker with other allocations.) Note that they don't account for your future income streams, but they should model the investable portfolio portion pretty accurately and you can think of extra income as reducing the money you need from your portfolio.

Withdrawal Rates -- finds the safe and perpetual withdrawal rates using standard spending assumptions
Retirement Spending -- lets you adjust the spending assumptions to see the effects of variable spending strategies

TL;DR: Don't get too caught up in precise numbers until you're confident that they actually apply to UK investors. Most SWR research is written for a US audience and includes several core assumptions that don't easily translate. Using tools designed for your home country can provide more accurate results and give you more confidence in your plan.

I hope that helps! That may not clearly answer your questions as simply as you hoped, but hopefully it points you in the right direction.
Topic Author
maloflora
Posts: 29
Joined: Sat Oct 10, 2009 4:27 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by maloflora »

Thanks Tyler! Praise from the master of portfoliocharts is praise indeed. And in fact I constructed my portfolio almost entirely on your site. It shows a roughly 6% real return but I'm being much more cautious than that in my assumptions given everything.

I can completely understand why SWR calculations would be affected by country, but for spreadsheets like VPW and TPAW surely they would self-correct? I.e. if the combination of UK returns and exchange rates meant my portfolio returned less than the expected rate in a given year, the next year I would just enter the new lower number and take out a bit less?

I agree local inflation is a bigger deal, particularly at the moment, but the long-term target is still 2% which is pretty close to what the calculators assume.

My worry is that if I start to second-guess either future returns or other macro-economic factors too heavily, I'll just end up making mistakes and going against the BH philosophy. Maybe I should try adjusting assumptions to see what the impact is.
User avatar
LadyGeek
Site Admin
Posts: 95696
Joined: Sat Dec 20, 2008 4:34 pm
Location: Philadelphia
Contact:

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by LadyGeek »

This thread is now in the Non-US Investing forum (UK withdrawal strategies). maloflora - this forum will attract members with non-US investing (and retirement) experience.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by Ben Mathew »

I think you are in good shape to retire and the ballpark withdrawal amounts are reasonable. You have clearly done a lot of work on this, and I'm glad that you found ABW and TPAW useful in your planning.

I modified the "TPAW with Monte Carlo Simulation -- for Retirees" spreadsheet to fit your retirement years and ran the simulation. Here's the spreadsheet if you want to try different inputs.

I'm assuming that the UK state pension of £9000 is inflation indexed as well. So the retirement income stream is

- £7,800 from age 50-59 (rental income)
- £45,000 from age 60-66 (add 37,200 from workplace pension)
- £54,000 from age 67-89 (add 9,000 UK state pension)

Real return of stocks and bonds per your assumptions:
- 4% stocks
- 1% bonds

A 34/66 asset allocation on the total portfolio translates to a 65/35 asset allocation on the savings portfolio. Since your current AA is 65% stock on your savings portfolio, I went with 34/66 on the total portfolio to match it.

g = 0% per OP. I think it's reasonable to start out with g=0 to compare with other strategies. But eventually that might be something you may want to try adjusting. More on that below.

With these inputs, you have £1,500,000 (current savings) + £1,293,487 (present value of pensions and rental income) = £2,793,487 total assets supporting retirement spending.

The Monte Carlo simulation with these inputs generates the following distribution of withdrawals:

Image

Age 50 withdrawal is £100,447 (£92,647 from portfolio + £7,800 from rent). At age 89, median withdrawal is still a healthy £89,519 but the 10th percentile is down to £55,403--just slightly above the rent+pensions income stream of £54,000. There were no liquidity problems at these percentiles (i.e. you did not run out of money before pensions start.)

If you want to reduce the chance of withdrawals dipping this low in old age and are willing to give up some consumption in early retirement (which it sounds like you might be), consider increasing g. The graph below shows the impact of increasing g to 0.75%. Age 50 withdrawal reduces to £88,235, but the median and 10th percentile withdrawal at age 89 increases to £106,855 and £64,507 respectively. Whether that's a good tradeoff depends on your preferences. But it's helpful to try different values for g and see this tradeoff explicitly. That's the benefit of Monte Carlo planning -- you can more clearly see the impact of your planning choices and make an informed decision.

Image

Let me know if you have any questions.
Last edited by Ben Mathew on Thu Oct 28, 2021 10:24 am, edited 1 time in total.
Total Portfolio Allocation and Withdrawal (TPAW)
longinvest
Posts: 5682
Joined: Sat Aug 11, 2012 8:44 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by longinvest »

From the VPW thread:
maloflora wrote: Wed Oct 27, 2021 3:03 pm I'm really attracted to this way of thinking about retirement spending, for reasons outlined in this thread: viewtopic.php?f=10&t=360739&p=6289993#p6289993

I wonder if I could please check that I'm ok to apply the VPW logic in the way I outline there as a UK investor? I can't see any reasons why if the inflation assumptions are very similar you couldn't just update portfolio performance annually as you recommend without having to make forecasts about UK rates of return separate from US ones.

I appreciate that logic doesn't work for SWR-based retirement strategies but I think I'm flexible enough (and have the base pension income) to avoid the need to follow that route.

Thanks in advance for any help!
Dear Maloflora,

Here are some comments.

maloflora wrote: Sat Oct 23, 2021 10:50 am I’m looking at retirement at age 50 with a 40-year timeframe
The VPW retirement approach doesn't rely on the retiree conveniently dying at (or before) a particular age for the plan to work.

maloflora wrote: Sat Oct 23, 2021 10:50 am The portfolio consists of 45% global equities, 20% UK equities, 20% UK gilts (government bonds), 10% gold and 5% cash
VPW is based on our investment philosophy. It was designed to be used with a globally-diversified balanced portfolio containing total-market stock and bond index funds or ETFs. I wouldn't know how to apply VPW to a portfolio containing speculative assets like gold.

maloflora wrote: Sat Oct 23, 2021 10:50 am VPW spreadsheet: £102,606 total spend in year 1
I guess that this is the estimated amount available for taxes and expenses. I see no mention of the VPW retirement worksheet's Required Flexibility which is a key element. Here's part of a post I wrote about it in the main VPW thread:
longinvest wrote: Mon Apr 12, 2021 8:08 pm It's important to correctly understand the meaning of the Required Flexibility estimate of the VPW Retirement Worksheet.

The worksheet's simulated -50% stock drop is a normal event for stocks. That's like saying that there will be a car accident in New York city tomorrow. It's not perfectly certain to happen, but it's nothing out of the ordinary. The portfolio could drop more during retirement. The VPW worksheet only aims to tell the retiree that a -50% stock drop shouldn't cause any change in level of comfort. When the portfolio starts to drop more, the retiree might have to start reducing comfort. But there must be comfort to start with, even after a -50% stock drop. If there's no such flexibility, the investor isn't ready to retire.

There's no floor to how low VPW withdrawals can get because there's no limit to how low future market returns (for both stocks and bonds) can get. That's the reality we have to live with. The future is uncertain and future returns are unknown. It's best to accept it.

When strictly needed, stable lifelong inflation-indexed income (in addition to Social Security and work pension, if any) can be bought from an insurance company in the form of a 2%-indexed SPIA*. It's expensive. The younger the retiree, the more expensive it gets. And it's illiquid. Liquidity is important.

* Single Premium Immediate Annuity.

I suggest reading the VPW forward test thread which illustrates how to use the VPW retirement worksheet. There's also a similar thread using the same worksheet but with a globally-diversified Canadian (instead of U.S.) portfolio in our Canadian sister forum.


I'll conclude by copying a post I wrote in January 2021:
longinvest wrote: Sat Jan 02, 2021 12:02 pm In life, we dress according to the current situation. We put on warm clothes, Winter boots, and a Winter coat when getting outside during the Winter. We dress more lightly during the Summer. We don't wait until Winter to go to the store in our Summer clothes to buy Winter clothes. We buy Winter clothes (and Summer clothes) in advance and keep both Winter and Summer clothes in our homes. Also, in the Summer, we don't put on a Winter coat; we take advantage of the warm weather, instead, to put on a swim suit and jump into the outside pool and have fun.

The VPW Retirement Worksheet works similarly. It allows us to withdraw money from the portfolio according to the current situation. When the portfolio balance is higher, withdrawals are bigger, and when the portfolio balance is lower, withdrawals are smaller. VPW doesn't dress up withdrawals for Winter during the Summer. In other words, it doesn't keep withdrawals small when the portfolio balance is high. But, it informs the retiree about the normal consequences of bad weather (rain, wind, cold) so that the retiree can prepare for it (e.g. it informs the retiree about the consequences of stocks losing -50%).
Best regards,

longinvest
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Topic Author
maloflora
Posts: 29
Joined: Sat Oct 10, 2009 4:27 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by maloflora »

Hi Ben,

Thank you so much for the comprehensive reply - it's really reassuring to know that my basics are broadly right. And yes, the UK state pension is fully index-linked (slightly controversially!).
Ben Mathew wrote: Wed Oct 27, 2021 6:12 pm g = 0% per OP. I think it's reasonable to start out with g=0 to compare with other strategies. But eventually that might be something you may want to try adjusting. More on that below.
The 'g' part of the formula is the bit I can't quite understand in this approach. As I understand it - crudely - VPW, ABW and TPAW all basically convert portfolios and future income streams back to a single NPV sum (the £2.8m you calculated) and then amortise that total portfolio over the retirement period to even out the ups and downs as various streams kick in. Different approaches have different ways of not running out of money in the mean time. So why would annual spending tend to fall, and need to be buoyed up by a positive g%? Is that an outcome of the backtesting? In the reality of putting it into practice would you just be updating the portfolio values and timeframe every year?

Sorry if I've misunderstood something crucial!
Topic Author
maloflora
Posts: 29
Joined: Sat Oct 10, 2009 4:27 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by maloflora »

Hi longinvest,

Thanks so much for the comprehensive reply. A few responses, if I may.
longinvest wrote: Thu Oct 28, 2021 12:15 am VPW is based on our investment philosophy. It was designed to be used with a globally-diversified balanced portfolio containing total-market stock and bond index funds or ETFs. I wouldn't know how to apply VPW to a portfolio containing speculative assets like gold.
I've had this level of gold in my portfolio for a long time and I find it stabilises the portfolio, particularly at times of economic shock. I've put this portfolio through a variety of backtests and it performs pretty well compared to traditional portfolios (and of course only 10% of the portfolio is non-traditional). Presumably it shouldn't make that much difference as it would be self-correcting over time, if returns turn out differently from the WAG?
longinvest wrote: Thu Oct 28, 2021 12:15 am I guess that this is the estimated amount available for taxes and expenses. I see no mention of the VPW retirement worksheet's Required Flexibility which is a key element... When the portfolio starts to drop more, the retiree might have to start reducing comfort. But there must be comfort to start with, even after a -50% stock drop. If there's no such flexibility, the investor isn't ready to retire.
I didn't draw this out specifically but I'm totally on board with this, and with the Winter/Summer analogy. It's why I said in the OP that I'm flexible around spending and an absolute minimum spend a year would be around £40,000 which is much less than the initial taxes and expenses figure of £105,000. In addition, the baseline two pensions bring me up to that anyway.

I hope that reassures you - hoping that that means I'm heading in the right direction with this combination of strategies.

Many thanks.
User avatar
NearlyRetired
Posts: 350
Joined: Fri Aug 09, 2019 12:56 pm
Location: United Kingdom

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by NearlyRetired »

Ben Mathew wrote: Wed Oct 27, 2021 6:12 pm I think you are in good shape to retire and the ballpark withdrawal amounts are reasonable. You have clearly done a lot of work on this, and I'm glad that you found ABW and TPAW useful in your planning.

I modified the "TPAW with Monte Carlo Simulation -- for Retirees" spreadsheet to fit your retirement years and ran the simulation. Here's the spreadsheet if you want to try different inputs.
Hi Ben

Thank you for doing this - it helped clarify my take on your sheet for my circumstances which are not disimilar to the OP. I do have one question about it, which I am unclear about.

On the copy that I took of this link, it looks like the figures in BN:UR are fixed(these I think feed into the Percentile graph) - is this correct or am I missing something - is there a step I should be doing?
To err is to be human, to really mess up, use a computer
Valuethinker
Posts: 49035
Joined: Fri May 11, 2007 11:07 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by Valuethinker »

Ben Mathew wrote: Wed Oct 27, 2021 6:12 pm
Real return of stocks and bonds per your assumptions:
- 4% stocks
- 1% bonds

These assumptions might be reasonable long term but they are not reasonable short term.

There's no sterling bond investment that I can make right now, net of default risk, that will give me 1% real return on bonds? Unless inflation drops to c -1% (gap w 10 year nominal gilt)? Indexed Linked Gilts are c -2.5%?

Stocks? Yes. Well. UK has lagged just about any other major index due to sectoral composition. The characterisation of the UK stock market as "an index of a former Empire" is broadly true. Maybe it catches up-- the world learns to love tobacco, drinks & petroleum again.

If we have the global index, then it's really a call about US stocks (60% ish of total equities). There, again, it's hard to see 4% real in the short term.

OP has some enormous advantages - a generous Final Salary pension & from age 60, not 65 or 67. That provides a lot of financial security, and the rental income helps a lot, too.

I would say to OP to plan on 0% real return on bonds for the next 10 years, and 3% real for stocks. But also to model say a -35% drop in stocks in the next 3 years, and c -2% real on bonds for the next 5 years. The old "Sequence of returns risk". Where do we think stock markets would be now if vaccines had not been released in December of 2020 and most of the world was still in lockdown**?

If we do get returns "fading" back to historic normal levels, that is more likely to occur due to a sharp correction in market values (a bear market in equities and possibly bonds) than it is via some gradual adjustment.

** rhetorical question as we are not allowed to discus Covid-19. Suffice it to say I think most people would agree markets would be lower now, if vaccine discovery had not been so wildly faster & better than in all of previous history of infectious disease?
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by Ben Mathew »

maloflora wrote: Thu Oct 28, 2021 4:06 am
Ben Mathew wrote: Wed Oct 27, 2021 6:12 pm g = 0% per OP. I think it's reasonable to start out with g=0 to compare with other strategies. But eventually that might be something you may want to try adjusting. More on that below.
The 'g' part of the formula is the bit I can't quite understand in this approach. As I understand it - crudely - VPW, ABW and TPAW all basically convert portfolios and future income streams back to a single NPV sum (the £2.8m you calculated) and then amortise that total portfolio over the retirement period to even out the ups and downs as various streams kick in. Different approaches have different ways of not running out of money in the mean time. So why would annual spending tend to fall, and need to be buoyed up by a positive g%? Is that an outcome of the backtesting? In the reality of putting it into practice would you just be updating the portfolio values and timeframe every year?

Sorry if I've misunderstood something crucial!
It comes down to the fact that the future is uncertain. When you set g=0, you are basically saying that you're planning to withdraw £100,000 today and £100,000 tomorrow. The £100,000 you withdraw today is certain. But the £100,000 you're scheduling for tomorrow is uncertain. If returns are higher than expected, you will get more. But if the returns are lower than expected, you will get less. This is why the withdrawals shown in the graph get more and more dispersed further in the future. The uncertainty increases with time. The distant future is less clear than the near future.

So if you care equally about consumption today and tomorrow, then you would want to give up a little bit today and get more tomorrow to make tomorrow safer. This is a natural instinct known as "precautionary savings." You displayed it when you wrote in the OP:
maloflora wrote: Sat Oct 23, 2021 10:50 am I should say that I am by nature a cautious soul, so it is highly unlikely I will spend to the limit of any calculator
What g does is model that caution. By trying out different g, you are basically evaluating how cautious you want to be.

Of course, there's the opposite effect as well--a desire to spend more in early retirement when you are physically able to travel and so on. That would mean going the other way and spending more early on. That can mean setting g<0. The final g is a balance between the two: the desire to consume more early and the urge to buy insurance for an uncertain future. The right balance is a personal choice. So the TPAW spreadsheet allows g<0 as well. But there is another option to consider: Make g>0 but schedule extra expenses in the early years in the extra expenses columns (M and O) to fund an active early retirement. This would work a little differently, and the spending profile generated this way might be more attractive to some people. Trying all of these options can help make an informed decision.
Last edited by Ben Mathew on Thu Oct 28, 2021 11:37 am, edited 1 time in total.
Total Portfolio Allocation and Withdrawal (TPAW)
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by Ben Mathew »

NearlyRetired wrote: Thu Oct 28, 2021 5:26 am
Ben Mathew wrote: Wed Oct 27, 2021 6:12 pm I think you are in good shape to retire and the ballpark withdrawal amounts are reasonable. You have clearly done a lot of work on this, and I'm glad that you found ABW and TPAW useful in your planning.

I modified the "TPAW with Monte Carlo Simulation -- for Retirees" spreadsheet to fit your retirement years and ran the simulation. Here's the spreadsheet if you want to try different inputs.
Hi Ben

Thank you for doing this - it helped clarify my take on your sheet for my circumstances which are not disimilar to the OP. I do have one question about it, which I am unclear about.

On the copy that I took of this link, it looks like the figures in BN:UR are fixed(these I think feed into the Percentile graph) - is this correct or am I missing something - is there a step I should be doing?
Those columns are the results of the 500 simulations. They will update only after you run the simulation. Follow the instructions in cell B5 to run the simulation:
After you have entered your inputs and are ready to run the Monte Carlo simulation, click on Formulas -> Calculate Now. This will take around 30 seconds or so, depending on the speed of your computer.
You should see those numbers update after the simulation runs.

Let me know if you need a spreadsheet to fit your retirement horizon. It doesn't take me long to modify one.
Total Portfolio Allocation and Withdrawal (TPAW)
User avatar
NearlyRetired
Posts: 350
Joined: Fri Aug 09, 2019 12:56 pm
Location: United Kingdom

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by NearlyRetired »

Ben Mathew wrote: Thu Oct 28, 2021 11:25 am Those columns are the results of the 500 simulations. They will update only after you run the simulation. Follow the instructions in cell B5 to run the simulation:
After you have entered your inputs and are ready to run the Monte Carlo simulation, click on Formulas -> Calculate Now. This will take around 30 seconds or so, depending on the speed of your computer.
You should see those numbers update after the simulation runs.

Let me know if you need a spreadsheet to fit your retirement horizon. It doesn't take me long to modify one.
Hmm, it's not for me BUT I am running this in LibreOffice (also tried in Google Sheets with same results) so I wonder if that is the reason why.... I don't wish to de-rail this thread so will PM you later if that is ok?
To err is to be human, to really mess up, use a computer
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by Ben Mathew »

NearlyRetired wrote: Thu Oct 28, 2021 11:42 am
Ben Mathew wrote: Thu Oct 28, 2021 11:25 am Those columns are the results of the 500 simulations. They will update only after you run the simulation. Follow the instructions in cell B5 to run the simulation:
After you have entered your inputs and are ready to run the Monte Carlo simulation, click on Formulas -> Calculate Now. This will take around 30 seconds or so, depending on the speed of your computer.
You should see those numbers update after the simulation runs.

Let me know if you need a spreadsheet to fit your retirement horizon. It doesn't take me long to modify one.
Hmm, it's not for me BUT I am running this in LibreOffice (also tried in Google Sheets with same results) so I wonder if that is the reason why.... I don't wish to de-rail this thread so will PM you later if that is ok?
I see. I am only familiar with Excel. So I don't know if it will work in LibreOffice or Google Sheets. But yes, PM me or post in the TPAW thread.
Total Portfolio Allocation and Withdrawal (TPAW)
Topic Author
maloflora
Posts: 29
Joined: Sat Oct 10, 2009 4:27 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by maloflora »

Ben Mathew wrote: Thu Oct 28, 2021 11:13 am What g does is model that caution. By trying out different g, you are basically evaluating how cautious you want to be.
Thank you! That's a really helpful explanation. I'll continue to experiment with different 'g's to balance current and future spending. g=0.5 seemed to come out with a pretty reasonable balance of today versus tomorrow.
Topic Author
maloflora
Posts: 29
Joined: Sat Oct 10, 2009 4:27 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by maloflora »

Valuethinker wrote: Thu Oct 28, 2021 5:42 am I would say to OP to plan on 0% real return on bonds for the next 10 years, and 3% real for stocks.
Thank you for the warning. I decided to rerun all the simulations using these parameters for the full 40 year period to test out a more gloomy outlook and to reflect the fact that UK returns have tended to undershoot US ones (though CAPE is currently a lot lower in the UK):
  • VPW spreadsheet: £102,606 total spend in year 1 - no change as the WAG shouldn't be changed. But as explained in another thread, I'm comfortable with spend levels under his downside scenario
  • Big ERN’s actuarial calculator using a 3% discount rate rather than 4%: £84,419
  • Siamond’s Amortized Spending Budget calculator: £89,963
  • Ken Steiner’s Actuarial Budget Calculator: £72,641, now assuming a 2.1% expected rate of return and 2% inflation
  • Ben Mathew’s Total Portfolio Allocation and Withdrawal Planner with Monte Carlo Simulation calculator: £100,084 (still assuming 0% growth in withdrawals)
  • Ben’s more simple Total Portfolio Allocation and Withdrawal Calculator: £97,290
  • Big ERN’s SWR calculator. Again setting, slightly arbitrarily, a 5% failure rate, I get a 5.16% safe withdrawal rate leading to a first year total spend of £85,200
What I find interesting is how little difference the r rate actually makes. For the PMT-based methods it seems to take off c£5-8k from the first year spend. That's partly to do with the fixed income streams but it's reassuring for this particular way of structuring withdrawals.

That's assuming I've got all this right again! Hope that all makes sense.
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by Ben Mathew »

maloflora wrote: Fri Oct 29, 2021 11:00 am
Ben Mathew wrote: Thu Oct 28, 2021 11:13 am What g does is model that caution. By trying out different g, you are basically evaluating how cautious you want to be.
Thank you! That's a really helpful explanation. I'll continue to experiment with different 'g's to balance current and future spending. g=0.5 seemed to come out with a pretty reasonable balance of today versus tomorrow.
Great. :thumbsup
Total Portfolio Allocation and Withdrawal (TPAW)
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by Ben Mathew »

maloflora wrote: Fri Oct 29, 2021 11:16 am
Valuethinker wrote: Thu Oct 28, 2021 5:42 am I would say to OP to plan on 0% real return on bonds for the next 10 years, and 3% real for stocks.
Thank you for the warning. I decided to rerun all the simulations using these parameters for the full 40 year period to test out a more gloomy outlook and to reflect the fact that UK returns have tended to undershoot US ones (though CAPE is currently a lot lower in the UK):
  • VPW spreadsheet: £102,606 total spend in year 1 - no change as the WAG shouldn't be changed. But as explained in another thread, I'm comfortable with spend levels under his downside scenario
  • Big ERN’s actuarial calculator using a 3% discount rate rather than 4%: £84,419
  • Siamond’s Amortized Spending Budget calculator: £89,963
  • Ken Steiner’s Actuarial Budget Calculator: £72,641, now assuming a 2.1% expected rate of return and 2% inflation
  • Ben Mathew’s Total Portfolio Allocation and Withdrawal Planner with Monte Carlo Simulation calculator: £100,084 (still assuming 0% growth in withdrawals)
  • Ben’s more simple Total Portfolio Allocation and Withdrawal Calculator: £97,290
  • Big ERN’s SWR calculator. Again setting, slightly arbitrarily, a 5% failure rate, I get a 5.16% safe withdrawal rate leading to a first year total spend of £85,200
What I find interesting is how little difference the r rate actually makes. For the PMT-based methods it seems to take off c£5-8k from the first year spend. That's partly to do with the fixed income streams but it's reassuring for this particular way of structuring withdrawals.

That's assuming I've got all this right again! Hope that all makes sense.
Reducing expected real return to 3% stocks and 0% bonds in the TPAW planner reduces age 50 withdrawal from £100,447 to £94,874. The relatively muted impact on your withdrawal is, as you said, because a substantial portion of your retirement expenses is being funded by your retirement income stream of rent+pensions.

Increasing g to 0.5% further reduces withdrawals to £86,519.

If the goal is to fully draw down the portfolio and maximize withdrawals, you may also want to consider transitioning to annuities at later ages. Not sure what the opportunities for annuitizing in the UK is, but that may be worth looking into at some point.
Total Portfolio Allocation and Withdrawal (TPAW)
Topic Author
maloflora
Posts: 29
Joined: Sat Oct 10, 2009 4:27 am

Re: Withdrawal strategy comparisons (adapting US to UK)

Post by maloflora »

Ben Mathew wrote: Fri Oct 29, 2021 3:56 pm If the goal is to fully draw down the portfolio and maximize withdrawals, you may also want to consider transitioning to annuities at later ages. Not sure what the opportunities for annuitizing in the UK is, but that may be worth looking into at some point.
Thank you. Yes that's definitely a possibility in the UK, though the rates are of course terrible at the moment. I had a bad experience with annuities with my mother (not the annuity's fault!) but they're definitely an option around age 80.
Post Reply