Safe Withdrawal Rate increase

newkeen

Has anyone read and understood the recent post "Ratcheting The Safe Withdrawal Rate For Income Upside - Kitces.com"?

It suggests that unless the aim is to leave a large amount unspent, the initial % withdrawal rate can be safely increased by 10% of the original amount each time a portfolio reaches 50% above its starting value, with no reduction needed in subsequent bad markets.

It suggests continuing to do this throughout retirement. The 4% refers to the US, and it could be 2% or 3% is better for the UK or anywhere else. But whatever the starting rate, how can you continue to use this approach. The portfolio will only cross the 50% above starting value once to trigger the increase withdrawal rate. So what happens after that? The portfolio may rise and fall above and below the starting value multiple times but I think only the initial crossing triggers a 10% increase. But this theory refers to retirees with growing portfolio values progressively increasing the withdrawal rate by 10% many times during retirement.

So how would you interpret this 50% threshold working?

Linton

I interpret it as....

Start at £100K and drawdown £4K annually. If/when your pot exceeds £150K increase the drawdown to £4.4K. Then when the portfolio exceeds £200K increase the drawdown to £4.84K etc etc. Never cut back. He does seem to say that the drawdowns are also incaresed with inflation so the figures given are at constant prices.

The whole article is rather long and confusing for what is a pretty simple concept. Jamesd has put together rather better information on the topic.

dunstonh

It suggests continuing to do this throughout retirement. The 4% refers to the US, and it could be 2% or 3% is better for the UK or anywhere else.

why would it be different?
asset class and risk taken will be the main driver and what is [STRIKE]safe[/STRIKE] sustainable.

It suggests that unless the aim is to leave a large amount unspent, the initial % withdrawal rate can be safely increased by 10% of the original amount each time a portfolio reaches 50% above its starting value, with no reduction needed in subsequent bad markets.

Broadly speaking makes sense. Although if you are drawing 4% and invest like most retired people, then it will be a long time before that happens.

So what happens after that? The portfolio may rise and fall above and below the starting value multiple times but I think only the initial crossing triggers a 10% increase.

What it is saying is that you should be able to increase the withdrawal once the value is 50% higher, so increasing the withdrawal at that point should make it safe as a crash could be 40%. That leaves a 10% increase still built in.

So how would you interpret this 50% threshold working?

You would need a high risk income portfolio and would need to be pretty lazy. You dont need to leave it like that. The concept is simple but the way the article is displaying it is overly complicating things.

System

Doesn't it depend on what you mean to do with your increased income?
If you simply add it mentally to your original target, and up your standard of living to enjoy it, it would be as if you had calculated you needed a bigger pot right from the start. In which case the same safety rules would apply at the higher level.

But if you continue to spend only your original target income, and take a more relaxed attitude to the bonus you now enjoy, then you could reduce your risk threshold without worry. If the worst came to the worst you might lose all the 50% increase, but you wouldn't mind because you would still have the income you had originally been counting on.

newkeen

dunstonh,

Thanks for the comments.

Interesting that you make the point that the 4% rule based on US historical data need not be different for UK investors.

Whenever I read about the 4% rule on UK investing websites it nearly always carries the caveat that returns for UK investors are historically lower and as a result the SWR is probably lower for UK investors. Is this based on the assumption that UK investors must be heavily weighted in UK stocks? But if the investor has a globally diversified portfolio with funds such as Vanguards global or Lifestyle funds is the difference less pertinent?


Linton,

Linton wrote: »

I interpret it as....

Start at £100K and drawdown £4K annually. If/when your pot exceeds £150K increase the drawdown to £4.4K. Then when the portfolio exceeds £200K increase the drawdown to £4.84K etc etc. Never cut back. He does seem to say that the drawdowns are also incaresed with inflation so the figures given are at constant prices.

Thank you Linton. I agree £100K to £150K is 50% so £4K becomes £4.4K. What I wasn't sure about was that the next increase point is still £50K (£200K) as you say, or 50% of the previous threshold (£150K/2 = £75K so £225K). You feel the article proposes it is always £50K increments?

Linton

newkeen wrote: »

dunstonh,


Thank you Linton. I agree £100K to £150K is 50% so £4K becomes £4.4K. What I wasn't sure about was that the next increase point is still £50K (£200K) as you say, or 50% of the previous threshold (£150K/2 = £75K so £225K). You feel the article proposes it is always £50K increments?

I dont think its clear, and to a great extent it doesnt matter that much - the difference between £225K and £200K isnt really significant as regards implementing the scheme. I think you should read the paper as a broad proposal rather than a strict recipe.

dunstonh

Interesting that you make the point that the 4% rule based on US historical data need not be different for UK investors.

Historically, 5% was considered safe in the UK until recently. It is part of the reason the old 5% deferment rule exited on life products.

However, the key thing about withdrawals is the underlying assets. If you have assets that are maybe consistent at 3% p.a. then a US quoted safe withdrawal rate of say 4% does not matter. It wouldnt be safe.

Whenever I read about the 4% rule on UK investing websites it nearly always carries the caveat that returns for UK investors are historically lower and as a result the SWR is probably lower for UK investors.

The UK FTSE100 is a woefully poor index and if you compared that with the S&P500 then they are miles apart. However, UK investors would not use a FTSE100 fund (unless very daft). So, you cant compare them like that. As I said, UK investors are generally more global in nature. Any article making out UK is worse for investors should put you on guard. The old multi-asset fund dominates use in the UK over any other type of fund. Has done for decades. This is not new to Vanguard.

You need to be wary when reading things aimed at US investors. UK taxation is more favourable on funds than in the US. Models for US investors are very US centric. You would not simply switch US large cap for UK large cap or US mid cap for UK mid cap.

newkeen

dunstonh wrote: »

Historically, 5% was considered safe in the UK until recently. It is part of the reason the old 5% deferment rule exited on life products.

However, the key thing about withdrawals is the underlying assets. If you have assets that are maybe consistent at 3% p.a. then a US quoted safe withdrawal rate of say 4% does not matter. It wouldnt be safe.


The UK FTSE100 is a woefully poor index and if you compared that with the S&P500 then they are miles apart. However, UK investors would not use a FTSE100 fund (unless very daft). So, you cant compare them like that. As I said, UK investors are generally more global in nature. Any article making out UK is worse for investors should put you on guard. The old multi-asset fund dominates use in the UK over any other type of fund. Has done for decades. This is not new to Vanguard.

You need to be wary when reading things aimed at US investors. UK taxation is more favourable on funds than in the US. Models for US investors are very US centric. You would not simply switch US large cap for UK large cap or US mid cap for UK mid cap.

Interesting that UK safe withdrawal rates were historically, higher, and could still be with UK tax advantages. But as you say, what really matters is the underlying assets. So how do you go about finding out what is a reasonable expectation for the assets you hold? In my case 70% equities, 20% bonds, 10% cash. Equities are global - Vanguard 100 Lifestyle, and Vanguard EM and Vanguard Small Caps. Bonds are also Vanguard global.

ex-pat_scot

newkeen wrote: »

dunstonh,

Thanks for the comments.

Interesting that you make the point that the 4% rule based on US historical data need not be different for UK investors.

Whenever I read about the 4% rule on UK investing websites it nearly always carries the caveat that returns for UK investors are historically lower and as a result the SWR is probably lower for UK investors. Is this based on the assumption that UK investors must be heavily weighted in UK stocks? But if the investor has a globally diversified portfolio with funds such as Vanguards global or Lifestyle funds is the difference less pertinent?

The SWR is based on the value of the underlying assets.
The "classic" study, giving the 4% figure, was based on the US S&P500 equities and bonds.
As you note, a similar exercise on UK gives a lower (3.5% to 3.7%) SWR, off FTSE returns. (It's not quite the FTSE100, as that wasn't a consistent measure way back, but close enough).


The message then, is if you are fully invested in UK then the historic returns are lower. If in US, then historic returns are higher.


Whilst your retirement funds will normally be drawn in £GBP, there is absolutely no reason why you need to have your underlying investments in the GB market.
Indeed, I think that is a spectacularly narrow / insular view of investment strategy.
Personally, a "whole world" broad tracker give a lot of geographic, political and FX diversification benefits and a greater long term return. That's why the vast majority of my investments are outside UK.

jamesd

newkeen wrote: »

Has anyone read and understood the recent post "Ratcheting The Safe Withdrawal Rate For Income Upside - Kitces.com"?

I hadn't read that particular one but have referred to the ability to raise the income level if returns are normal before. The 4% plus ramps up in income mean that you'll end up getting the higher income in older age when you're perhaps least able to use it.

One reason I like Guyton and Klinger's rules is that they do some increasing but it's still not as flexible on the upside as it might be. And there's the option in there of having a spending floor. They also let you start higher than 4%, at an age when you might be better able to spend, and with cfiresim you can build in variable spending patterns as well as part of the plan, by adding fake "income" of the amount you want income to decrease by.

newkeen wrote: »

It suggests that unless the aim is to leave a large amount unspent, the initial % withdrawal rate can be safely increased by 10% of the original amount each time a portfolio reaches 50% above its starting value, with no reduction needed in subsequent bad markets.

Rather than doing that I'd just use the original calculations for the remaining life expectancy and work out the new safe withdrawal rate. But not a drop, an increase only review, because the spending rules already protect against the drop case.

newkeen wrote: »

So what happens after that? The portfolio may rise and fall above and below the starting value multiple times but I think only the initial crossing triggers a 10% increase. But this theory refers to retirees with growing portfolio values progressively increasing the withdrawal rate by 10% many times during retirement. ... So how would you interpret this 50% threshold working?

The reason you can do that is that the rule being used works at any time, so if you do well you can increase the income to whatever the rule says meets your confidence level target. And barring the worst case scenarios you will be able to do that.

jamesd

Anyone interested in this will probably like these two videos, I suggest the bottom one with Guyton, of the Guyton and Klinger rules, first: Guyton and Kitces with Continued Discussions on Safe Withdrawal Rates and Spending Decision Rules