Your browser isn't supported
It looks like you're using an old web browser. To get the most out of the site and to ensure guides display correctly, we suggest upgrading your browser now. Download the latest:

Welcome to the MSE Forums

We're home to a fantastic community of MoneySavers but anyone can post. Please exercise caution & report spam, illegal, offensive or libellous posts/messages: click "report" or email forumteam@.

Search
  • FIRST POST
    • michaels
    • By michaels 17th May 19, 11:13 PM
    • 22,514Posts
    • 103,516Thanks
    michaels
    SWR start point and market value on retirement
    • #1
    • 17th May 19, 11:13 PM
    SWR start point and market value on retirement 17th May 19 at 11:13 PM
    So I imagine stock markets as following a slight upward line over time with lots of peaks and trough noise either side of the trend line.

    If this is reasonable as a model then does that mean on retirement rather than taking say 4% of your ppt value on retirement you should take 4% of the trend fit line at that point. Eg if the market is 10% above trend then your starting withdrawal rate for whichever model you use should be based on a pot 10% smaller and vice versa if the market is below long term trend.

    Are there any papers discussing this sort of approach to drawdown?

    Thanks
    Cool heads and compromise
Page 1
    • Marcon
    • By Marcon 17th May 19, 11:53 PM
    • 1,028 Posts
    • 788 Thanks
    Marcon
    • #2
    • 17th May 19, 11:53 PM
    • #2
    • 17th May 19, 11:53 PM

    Are there any papers discussing this sort of approach to drawdown?
    Originally posted by michaels
    No because it isn't really a feasible or sensible approach to take, given the unpredictable nature of the markets (especially in recent years) - and of course much depends on where your funds are invested.
  • jamesd
    • #3
    • 18th May 19, 1:25 AM
    • #3
    • 18th May 19, 1:25 AM
    If this is reasonable as a model then does that mean on retirement rather than taking say 4% of your ppt value on retirement you should take 4% of the trend fit line at that point. Eg if the market is 10% above trend then your starting withdrawal rate for whichever model you use should be based on a pot 10% smaller and vice versa if the market is below long term trend.

    Are there any papers discussing this sort of approach to drawdown?
    Originally posted by michaels
    Yes, to a degree, though you have it somewhat the wrong way around. The SWR is based on the worst case so if you see that markets are in good shape - down! - you can expect a higher withdrawal percentage to work. Three sets of writing to consider are:

    1. Guyton's sequence of returns risk taming approach, which varies asset allocation based on the cyclically adjusted price/earnings ratio.
    2. Kitces work explaining the high inverse correlation between cyclically adjusted P/E and ten year returns, some at the 1 link. Also Shiller CAPE Market Valuation: Terrible For Market Timing, But Valuable For Long-Term Retirement Planning and Resolving the Paradox - Is the Safe Withdrawal Rate Sometimes Too Safe? He observed that Shiller CAEP (E/P not P/E) had a high 0.77 correlation with the 30 year SWR.
    3. Wade Pfau has also done work in this area and his US dashboard may be interesting. Scroll down to (2) the penultimate table for the summary. The aggressive column seems the one that best matches traditional SWR calculations.

    These pieces of work, notably 1, are a large part of why I currently suggest considering lower equity or lower US percentage in the first post of Drawdown: safe withdrawal rates.
    Last edited by jamesd; 18-05-2019 at 1:28 AM.
    • MK62
    • By MK62 18th May 19, 6:24 AM
    • 453 Posts
    • 327 Thanks
    MK62
    • #4
    • 18th May 19, 6:24 AM
    • #4
    • 18th May 19, 6:24 AM
    So I imagine stock markets as following a slight upward line over time with lots of peaks and trough noise either side of the trend line.

    If this is reasonable as a model then does that mean on retirement rather than taking say 4% of your ppt value on retirement you should take 4% of the trend fit line at that point. Eg if the market is 10% above trend then your starting withdrawal rate for whichever model you use should be based on a pot 10% smaller and vice versa if the market is below long term trend.
    Originally posted by michaels

    As your topic title is about "Safe Withdrawal Rate", then assuming your pot is smaller than it actually is (if the markets are above this trend line) could be seen as building in some safety margin, if the resulting size of withdrawal is still viable for you.

    However, assuming your pot is bigger than it actually is (if the markets are below this trend line) could equally be seen as building in a "danger" margin, and exposing you to higher sequence of return risk right at the start, exactly when you don't want it. The markets could take years to recover back to the trend line, and even then it may only be because the lower markets then alter the trend line downwards as the years go by.
    • Sea Shell
    • By Sea Shell 18th May 19, 6:39 AM
    • 1,939 Posts
    • 3,313 Thanks
    Sea Shell
    • #5
    • 18th May 19, 6:39 AM
    • #5
    • 18th May 19, 6:39 AM
    We're working on the principal (rightly or wrongly) that we need to achieve 3% from our starting baseline pot. This will meet our usual spending needs without reducing capital. (this excludes SP when they kick in....eventually!!)

    At present our pot stands at £24,000 above our baseline, due to recent fluctuations, so we have some "flexibility" built in already.

    Obviously if we increased our spending a bit, or made any large purchases, it all has to come out of the pot, so we'd review as we go, in that respect. The same with inflation, if it started to get out of control, relative to our remaining pot at any given time, we'd reassess our plan.

    IMO I think flexibility is the key. Being able to (or wanting to) adjust your lifestyle depending on current market conditions (within reason), so making hay whilst the sun shines, but being prepared to reign it in if the wheels look like they're falling off!!! - But that's just us!!!
    " That pound I saved yesterday, is a pound I don't have to earn tomorrow "
    • saver_ali
    • By saver_ali 18th May 19, 7:41 AM
    • 76 Posts
    • 32 Thanks
    saver_ali
    • #6
    • 18th May 19, 7:41 AM
    • #6
    • 18th May 19, 7:41 AM
    I'm currently reading a book that was recommended on an earlier thread, called "Beyond the 4% rule" by Abraham Okusanya, which claims to discuss "the science of retirement portfolios that last a lifetime". It's good so far, but I'm only a third of the way through it!
    • Thrugelmir
    • By Thrugelmir 18th May 19, 8:05 AM
    • 63,477 Posts
    • 56,285 Thanks
    Thrugelmir
    • #7
    • 18th May 19, 8:05 AM
    • #7
    • 18th May 19, 8:05 AM
    So I imagine stock markets as following a slight upward line over time with lots of peaks and trough noise either side of the trend line.
    Originally posted by michaels
    Are you referring to the market indices?

    You make no reference to a fixed interest portfolio either. A 100% equity exposure could swing wildly depending on the markets that you invest in. Markets in themselves being a very broad term.
    Last edited by Thrugelmir; 18-05-2019 at 8:07 AM.
    “The stock market is a device for transferring money from the impatient to the patient.” – Warren Buffett
    • Linton
    • By Linton 18th May 19, 8:42 AM
    • 10,730 Posts
    • 11,103 Thanks
    Linton
    • #8
    • 18th May 19, 8:42 AM
    • #8
    • 18th May 19, 8:42 AM
    So I imagine stock markets as following a slight upward line over time with lots of peaks and trough noise either side of the trend line.

    If this is reasonable as a model then does that mean on retirement rather than taking say 4% of your ppt value on retirement you should take 4% of the trend fit line at that point. Eg if the market is 10% above trend then your starting withdrawal rate for whichever model you use should be based on a pot 10% smaller and vice versa if the market is below long term trend.

    Are there any papers discussing this sort of approach to drawdown?

    Thanks
    Originally posted by michaels

    This seems wrong for two reasons, one simple and the other rather deeper.....

    A) A trend line is simply a rationalisation of the past. It gives you no information as to the future. The danger is that it could easily lead you to taking too much money too early on.

    B) In my experience the model you use for planning retirement does not and should not be used for what you actually drawdown during retirement. Its sole purpose is to give you the confidence to stop working and to start living off your savings/investments. Within a very few years you are likely to find that some of the assumptions you made in your planning no longer apply. At that point the logical response is to make a new long term plan, not to carry on blindly with the old one.
    • AnotherJoe
    • By AnotherJoe 18th May 19, 9:03 AM
    • 14,657 Posts
    • 17,505 Thanks
    AnotherJoe
    • #9
    • 18th May 19, 9:03 AM
    • #9
    • 18th May 19, 9:03 AM
    Without seeing any numbers £24k seems very close to the bone to me..but 3% is also a lowball number so that compensates somewhat. As, frankly is not allowing the pot to decrease when you've also got (presumably) 2x SP coming on stream soonish which will mean you can decrease the 3% take
    Please dont criticise my spelling. It's excellent. Its my typing that's bad.
    • Sea Shell
    • By Sea Shell 18th May 19, 9:08 AM
    • 1,939 Posts
    • 3,313 Thanks
    Sea Shell
    Without seeing any numbers £24k seems very close to the bone to me..but 3% is also a lowball number so that compensates somewhat. As, frankly is not allowing the pot to decrease when you've also got (presumably) 2x SP coming on stream soonish which will mean you can decrease the 3% take
    Originally posted by AnotherJoe
    Sorry, were you referring to my £24k?? I didn't post my actual pot numbers, just that we're that much over our benchmark at present, so therefore a drop of £24k would not affect our plan. IYSWIM.
    " That pound I saved yesterday, is a pound I don't have to earn tomorrow "
  • jamesd
    Without seeing any numbers £24k seems very close to the bone to me..but 3% is also a lowball number so that compensates somewhat. As, frankly is not allowing the pot to decrease when you've also got (presumably) 2x SP coming on stream soonish which will mean you can decrease the 3% take
    Originally posted by AnotherJoe
    This would normally substantially increase the initial take to cover for the missing state pensions. Then fall back based on the resulting lower pot. Or as Guyton does in his practice, a split:

    1. bridging pot to cover the missing pensions
    2. lifetime safe withdrawal rate income from the rest, or combination of some lifetime and some discretionary.
    • michaels
    • By michaels 18th May 19, 10:20 AM
    • 22,514 Posts
    • 103,516 Thanks
    michaels
    Thanks all for the reading suggestions

    No because it isn't really a feasible or sensible approach to take, given the unpredictable nature of the markets (especially in recent years) - and of course much depends on where your funds are invested.
    Originally posted by Marcon
    Imagine the following scenarios for retiring in 12 m time, pot currently worth 900k: (obviously extremely simplified for example purposes)
    1 'markets' and pension pot are up by 10% so a 900k pot is now about 1m, 4% swr is 40k pa
    2 'markets' down 10% so pot now worth 800k, 4% swr gives 32k pa.

    To me the likelihood of failure is higher in option 1 than option 2 but the swr model suggests otherwise?

    This seems to sum it up
    https://www.kitces.com/may-2008-issue-of-the-kitces-report/
    Tldr: markets high (on a 10 year rolling pe basis) swr 4.4%, middling 4.8%, low 5.7%
    Last edited by michaels; 18-05-2019 at 11:17 AM.
    Cool heads and compromise
    • Triumph13
    • By Triumph13 18th May 19, 10:46 AM
    • 1,465 Posts
    • 1,998 Thanks
    Triumph13
    Thanks all for the reading suggestions



    Imagine the following scenarios for retiring in 12 m time, pot currently worth 900k: (obviously extremely simplified for example purposes)
    1 'markets' and pension pot are up by 10% so a 900k pot is now about 1m, 4% swr is 40k pa
    2 'markets' down 10% so pot now worth 800k, 4% swr gives 32k pa.

    To me the likelihood of failure is higher in option 1 than option 2 but the swr model suggests otherwise?
    Originally posted by michaels
    It would seem so at first sight, but not necessarily as markets aren't just bouncing around while following an underlying trend. In your first case I'm not sure there is any historic data to show that this situation means you're more likely to be at the top of a bull run than in the start or middle or a period of sustained growth. Similarly case 2 could as easily be the start of a prolonged slump as a temporary blip.
    Personally I think there is also a distinction to be drawn between what the maths based on past history says and what helps you sleep at night. I'm a bad sleeper so I use the lower of my forecast investment number and actuals and apply my percentages to that. YMMV.
    • michaels
    • By michaels 18th May 19, 11:16 AM
    • 22,514 Posts
    • 103,516 Thanks
    michaels
    It would seem so at first sight, but not necessarily as markets aren't just bouncing around while following an underlying trend. In your first case I'm not sure there is any historic data to show that this situation means you're more likely to be at the top of a bull run than in the start or middle or a period of sustained growth. Similarly case 2 could as easily be the start of a prolonged slump as a temporary blip.
    Personally I think there is also a distinction to be drawn between what the maths based on past history says and what helps you sleep at night. I'm a bad sleeper so I use the lower of my forecast investment number and actuals and apply my percentages to that. YMMV.
    Originally posted by Triumph13
    Thanks.

    The paper I quote above has done the maths and the best predictor of medium term returns and thus swr is the 10 year rolling pe ratio. Top quintile (and thus all period) swr in their example is 4.4%, bottom quintile is 5.7% and mid quintiles 4.8%. Thus a small but real effect.

    However as you say efficient market hypothesis is that past performance should not be a future determinant and so just because a relationship has been observed does not mean it will continue...but then the same is also true of the 4% swr rule....
    Cool heads and compromise
    • AnotherJoe
    • By AnotherJoe 18th May 19, 11:24 AM
    • 14,657 Posts
    • 17,505 Thanks
    AnotherJoe
    Sorry, were you referring to my £24k?? I didn't post my actual pot numbers, just that we're that much over our benchmark at present, so therefore a drop of £24k would not affect our plan. IYSWIM.
    Originally posted by Sea Shell

    My thought was that if £24k over your benchmark is of note then it could easily swing the other way and you could quickly move to £24k under. Would that affect your plan?
    This is based on the thought that if you were say £100k over and withdrawing say £20k you wouldnt even think to ask, so you must be somewhere close to the limits in % term?
    Please dont criticise my spelling. It's excellent. Its my typing that's bad.
    • Thrugelmir
    • By Thrugelmir 18th May 19, 12:41 PM
    • 63,477 Posts
    • 56,285 Thanks
    Thrugelmir
    This seems to sum it up
    https://www.kitces.com/may-2008-issue-of-the-kitces-report/
    Tldr: markets high (on a 10 year rolling pe basis) swr 4.4%, middling 4.8%, low 5.7%
    Originally posted by michaels
    May 2008 was a different era. Some rather major financial events happened shortly afterwards and have continued since.

    By markets you appear to be referring to US equities and treasury stocks. Without going into depth, current levels of interest rates on bonds and exposure to currency movements are obvious question marks on a forward looking basis. If you propose to mirror such portfolios. Then there's the fundamentals such as p/e ratios etc to be considered as well.
    “The stock market is a device for transferring money from the impatient to the patient.” – Warren Buffett
    • michaels
    • By michaels 18th May 19, 1:07 PM
    • 22,514 Posts
    • 103,516 Thanks
    michaels
    May 2008 was a different era. Some rather major financial events happened shortly afterwards and have continued since.

    By markets you appear to be referring to US equities and treasury stocks. Without going into depth, current levels of interest rates on bonds and exposure to currency movements are obvious question marks on a forward looking basis. If you propose to mirror such portfolios. Then there's the fundamentals such as p/e ratios etc to be considered as well.
    Originally posted by Thrugelmir
    I'm not planning on a specific portfolio or rule going forward, I was just observing that based on historic experience (pre 2008) the data suggests that the achievable swr was lower when the retirement date was at a point when the 'markets' were 'high' by some metric and higher when it was 'low'.

    By that metric markets are currently very high. The models also show 60:40 equity bonds as producing the highest swrs. As you point out real bond returns are now much lower than historically (perhaps for the reasons in my next paragraph) so this alone would put a drag on any future swr.

    However all these backward looking models cover periods when the demographics were conducive to high levels of investment and growth, of course internationally to some extent that is still true but we are gradually moving to a position where the majority of the world is no longer in a position where the dependency ratio will drive growth.
    Cool heads and compromise
    • Thrugelmir
    • By Thrugelmir 18th May 19, 1:46 PM
    • 63,477 Posts
    • 56,285 Thanks
    Thrugelmir
    I'm not planning on a specific portfolio or rule going forward, I was just observing that based on historic experience (pre 2008) the data suggests that the achievable swr was lower when the retirement date was at a point when the 'markets' were 'high' by some metric and higher when it was 'low'.
    Originally posted by michaels
    The data is entirely US centric that's my point. You appear to be trying to find evidence to substantiate a view. Rather than maintaining an open mind.
    “The stock market is a device for transferring money from the impatient to the patient.” – Warren Buffett
    • westv
    • By westv 18th May 19, 2:12 PM
    • 4,814 Posts
    • 2,403 Thanks
    westv
    Personally I will probably base any percentage on my pots highest value achieved whether it is prior to retirement or at the point of retirement.
    • Sea Shell
    • By Sea Shell 18th May 19, 2:20 PM
    • 1,939 Posts
    • 3,313 Thanks
    Sea Shell
    At the end of the day, we're all just guessing / hoping what will happen in the future, roughly based on what's been before. We don't have crystal balls!!

    If you worried too much about the sustainability of your income/lifestyle, you'd never retire. Just keep working until you drop, where's the fun in that!!
    " That pound I saved yesterday, is a pound I don't have to earn tomorrow "
Welcome to our new Forum!

Our aim is to save you money quickly and easily. We hope you like it!

Forum Team Contact us

Live Stats

203Posts Today

2,631Users online

Martin's Twitter