Drawdown: safe withdrawal rates

jamesd

stoozie1 wrote: »

Do you mean you will be doing it in 18/19 for 17/18, or can you apply it to 2 tax years ago?

Thanks, corrected to : "For 2017/18 tax year I didn't buy VCTs. Instead I plan to buy EIS' in 2018/19 and apply that to the 2017/18 year"

You can go back to the previous tax year under EIS rules but two years isn't allowed.

Reluctantpensioner

Firstly, thanks everyone, especially jamesd, for this fascinating and articulate thread.

It encouraged me to read - a lot. I recently read many of the posts on Kitces site.
Several of his articles discuss linking SWR with a general market forecast.
His arguments makes a lot of sense.

Specifically, using the measure PE10 as a forecast of future market performance, he proposes derating SWR if PE10 is high (the market is overvalued) and increasing if low.
He has this table (sorry, can't make the formatting stick):
Market PE10 Recommendation
Overvalued >20 SWR=4.5%
neutral 12-20 SWR+0.5% = 5%
Undervalued <12 SWR+1% = 5.5%

A later article also proposes reducing equity:bond ratio when the market is overvalued and vice versa:
PE10 Recommendation
>18 Equity - 20%
<15 Equity +20%

MAJOR EDIT>> my post lost a whole section. All the questions were left out.
I note that US PE10 is 31.8 - well into overvalued. But I live in the UK...
For the UK I find FTSE100 15.22 and FTSE250 24.03. Two pretty different measures, apparently saying different things.
I normally prefer wider measures rather than concentrated ones. Is the UK overvalued or not?
Or is PE10 a relative measure, and I can't compare them like this? Gut feel says it's an absolute measure.
So that begs the question: what would trigger levels like Kitces be for the UK?
And of course there is the general question: what do you guys think of this approach?

jamesd

I've updated the first post to say a bit more about what to expect and not expect from cyclically adjusted P/E, since anyone who doesn't understand it might be puzzled by why I've been pointing to places saying now is a better time for lower equity holdings than high for a while.

Anonymous101

This is a very interesting article for anyone interested in drawdown rates and in particular early retirement.

https://www.gocurrycracker.com/the-worst-retirement-ever/

jamesd

Reluctantpensioner wrote: »

I note that US PE10 is 31.8 - well into overvalued. But I live in the UK...
For the UK I find FTSE100 15.22 and FTSE250 24.03. Two pretty different measures, apparently saying different things.
I normally prefer wider measures rather than concentrated ones. Is the UK overvalued or not?
Or is PE10 a relative measure, and I can't compare them like this? Gut feel says it's an absolute measure.
So that begs the question: what would trigger levels like Kitces be for the UK?
And of course there is the general question: what do you guys think of this approach?

Cyclically adjusted P/E needs to be calculated individually for each market. Each has its own long term average P/E and the degree to which they vary around it also varies. So it is more of a relative value than an absolute one. One of the earlier posts links to data for a range of markets.

If you're using a global tracker you could start with a cyclically adjusted P/E for that but then go on to look at the biggest constituent markets. If you think that one is high and the others aren't or that most are high but some aren't you could consider buying a short ETF to reduce your effective holding in the overvalued ones or add some extra outside holdings of the ones that look relatively good. So shifting your own effective average cyclically adjusted P/E down.

I think that cyclically adjusted P/E should be used because it does have preductive power and can reduce or eliminate the desire to reduce the safe withdrawal rate used. Read the Guyton paper.

Use the cyclically adjusted P/Es for the places where your investments live, not where you live. Safe withdrawal rates are normally calculated assuming residence in a country and all investments being made in that country, no diversification outside it. Sometimes they are calculated for global, though, and for the UK that actually produced a slightly lower constant inflation adjusted safe withdrawal rate.

It's also imporant to recognise that safe withdrawal rates are safe or unsafe depending on the rules used. 5% would be high for the 4% rule's level inflation adjusted income but it's the initial safe withdrawal rate for the UK if the Guyton-Klinger rules are used. Initial being the key word: unlike the 4% rule G-K adjusts up and down based on actual performance that you live through.

You're also free to restart your safe withdrawal rate calculation whenever you like and it is particulay important to do that for inflexible rules like the 4% rule if you find that you've lived through many bad years or are starting at a time of high cyclically adjusted P/E. Or if it has been many good years and you would find higher income useful.

Sucess rates also matter. For countries defeated, occupied or highly spending during WW2 the safe withdawal rate is reduced by war costs and damage, while it wasn't enough harm in the US to set the worst case rate that a 100% success rate has to use. It was for the UK where the.worst case for 100% sucess requirement is retiring a little before the war. You'd notice being in a country engaged in such a war and could adjust down or perhaps leave so it seems sensible to use a 95% or 90% success rate instead of 100%. Provided you're willing to adjust if those really bad things happen.

Success rate that you choose to use can also be lowered if you're highly flexible or have much of your essential income guaranteed by say the state pension, final salary pension or some annuities. Those reduce the worst case consequences if you do happen to live through bad investing times.

As you work through the links in the initial posts you'll find links to study and discussion of the things I've highlighted in this poat so that's some more homework to do.

DairyQueen

James, could you answer a quick question please? and apologies if it's a stupid one.

I am working from a base equity allocation of 65%. Should I include our cash holdings as an asset class when calculating/rebalancing the portfolio based on the PE10 rules?. I ask as we are planning on holding between 2 and 3 years of income in cash, and whether it is included affects the percentage of equities in the overall portfolio.

kidmugsy

jamesd wrote: »

I've updated the first post

I've not seen you posting here for a while, jamesd. Welcome back.

jamesd

DairyQueen wrote: »

James, could you answer a quick question please? and apologies if it's a stupid one.

I am working from a base equity allocation of 65%. Should I include our cash holdings as an asset class when calculating/rebalancing the portfolio based on the PE10 rules?. I ask as we are planning on holding between 2 and 3 years of income in cash, and whether it is included affects the percentage of equities in the overall portfolio.

I do, cash is an asset class. I happen to have a lot of it at the moment, way more than a few years of spending, but I have some fairly expensive work planned and don't mind being higher in cash for a while in current market conditions. Nothing horribly wrong with being high cash for a while outside raging bull market times.

Asking beats not asking.

Addendum: Guyton mentioned that cash should be counted as part of the bond percentage.

DairyQueen

jamesd wrote: »

I do, cash is an asset class. I happen to have a lot of it at the moment, way more than a few years of spending, but I have some fairly expensive work planned and don't mind being higher in cash for a while in current market conditions. Nothing horribly wrong with being high cash for a while outside raging bull market times.

Asking beats not asking.

Thanks for the quick response. I had been excluding unwrapped cash from the calculation so was considering selling more equities than I needed to.

I had a look through the early posts on this thread but am struggling to find the PE10 values for markets other than the US (SP 500) and UK (FTSE 100 and 250). Echoing the enquiry made upthread, do you know of any online source that lists the PE10 for all major indexes, by region, in one place?

(Asking beats not asking every time )

coastline

DairyQueen wrote: »

Thanks for the quick response. I had been excluding unwrapped cash from the calculation so was considering selling more equities than I needed to.

I had a look through the early posts on this thread but am struggling to find the PE10 values for markets other than the US (SP 500) and UK (FTSE 100 and 250). Echoing the enquiry made upthread, do you know of any online source that lists the PE10 for all major indexes, by region, in one place?

(Asking beats not asking every time )

This site might help.

https://www.gurufocus.com/global-market-valuation.php?country=GBR

https://www.gurufocus.com/shiller-PE.php

A question ? What do you do when markets are on reasonable forward P/E's ?

https://www.ceicdata.com/en/indicator/united-kingdom/pe-ratio

https://1.bp.blogspot.com/-UfijCaXv3JA/W-8wmxh93WI/AAAAAAAAr7A/VDVEjw8LgsgWI0U-KWaakb0DI6YChp2eACLcBGAs/s1600/PE%2B2.png

or this ?

https://www.youtube.com/watch?v=896xQ5qvBlc&feature=youtu.be&t=19m13s

https://www.youtube.com/watch?v=Sfg8J2jdyDU