When Does True SWR Begin?

[big-papa]

So big-papa, as an example if one started with 1mm in 1966 using 4% of remaining portfolio with the 5% real adjustment as necessary, what would be the portfolio after 30 years?
I always appreciate your WR% posts.

Good question - I'll pull out my backtester-o-matic today or tomorrow and then post the results... I'll do it for a few typical AA's.

Cheers!

 

[Oakster]

Ok.

My musing is about at what point does one declare themselves 'retired' whereupon they'd say: "Ok, now what is X% of our portfolio that we'll withdraw every year and adjust for inflation for the next 20 years; that number starts today!" even though they might have been withdrawing smaller amounts for years.

It seems that the difference is more about determining what that number is versus actually putting the 'X% plus inflation' to actual practice.

Personally, I wouldn't withdraw anymore than I needed or wanted to spend. I would want to have a 3 or 4 year CD ladder established to live off of is the market was down so I would not have to harm my nest egg in a downturn. If I had that in place, I would withdraw what I needed to live my desired lifestyle or I reached that number. If my wife is still working, I wouldn't need that much. If I had a pension, SS or a severance, I wouldn't need that much. Letting the nest egg grow early in retirement is a pretty huge benefit. At that point, your goals tell you what to do. If you want to leave an inheritance (legacy) draw what you need to live your desired lifestyle and stop at the chosen WR. If you want to spend it all, you can withdraw up to the WR and live higher than you need.

 

[Dtail]

Good question - I'll pull out my backtester-o-matic today or tomorrow and then post the results... I'll do it for a few typical AA's.

Cheers!

Thanks @big-papa. Appreciate it.

 

[audreyh1]

Thanks for that link. I downloaded the paper too. What I may have missed in skimming the 47 pages is any simple plot of portfolio value versus the various withdrawal rates.

But the web page does seem to show what I was talking about - the adjustable withdrawal approaches have you cutting significantly (more than ~ 25% cut in spending from 4%) for a decade or more, and those dips in WR go down to below 2.4% ( ~ 60% cut from 4%). That's a long time to do without so many things that could enhance your life.

To each their own, but I prefer to plan for a conservative WR of ~ 3.2%. And since it is conservative, I would also consider increasing that if the portfolio grew faster than inflation, and since I could also adjust for lower life expectancy years remaining.

-ERD50

Showing the actual real drawdown of the portfolio over the time period as well as the annual income variation is really important otherwise you are only seeing half the picture. It doesn’t seem to be there.

 

[big-papa]

Thanks @big-papa. Appreciate it.

OK. Here you go. It turns out I had this analysis stuffed away already. So here's the setup:
Data is from the Simba Spreadsheet over on Bogleheads
This is a 60/40 Portfolio with 60% Total Stock Market (VTSMX) and 40% Intermediate Treasuries. The difference won't be much if you used an SP500 fund and/or a total bond fund.
Clyatt is: 4% of the remaining nominal portfolio or 95% of the previous year's nominal withdrawal, whichever is larger.
Clyatt_Mod is: 4% of the remaining nominal portfolio or 95% of the previous year's real withdrawal, whichever is larger.
First plot is the real withdrawals for both methods.
Second plot is the nominal Remaining portfolio.
Because I had already had this built up, I did it for 40 years.
Results are as I discussed before. In real terms, using 95% real, definitely smooths things over during this timeframe, but there is a small penalty in that you're withdrawing more from your portfolio.

At 30 years, the remaining portfolio for Clyatt is: $11.7M
At 30 years, the remaining portfolio for Clyatt_Mod is: $10.2M

Of course the spending power from both portfolios is significantly less by that time.

For Clyatt, the minimum real withdrawal is: $43.6K
For Clyatt_Mod, the minimum real withdrawal is: 47.7K

The minimum portfolio value for Clyatt is: $2.02M
The minimum portfolio value for Clyatt_Mod is: $1.99M

For withdrawals starting in 1966, Clyatt was back to $100K by 1997
For withdrawals starting in 1966, Clyatt_Mod was back to $100K by 1998

The charts:

 

[Dtail]

Thanks @big-papa. Very helpful.
Couple of questions/points:
Am I reading it correctly, in that the starting point is 2mm with a 100k withdrawal, which would be a WR% of 5% instead of 4%?
Sounds like either methodology has a great survival rate and keeping the original principle, but at the expense of reducing withdrawals down by over 50% at one point and taking 30 years to get the spending made whole again.
A little scary analysis overall.

 

[big-papa]

Thanks @big-papa. Very helpful.
Couple of questions/points:
Am I reading it correctly, in that the starting point is 2mm with a 100k withdrawal, which would be a WR% of 5% instead of 4%?
Sounds like either methodology has a great survival rate and keeping the original principle, but at the expense of reducing withdrawals down by over 50% at one point and taking 30 years to get the spending made whole again.
A little scary analysis overall.

Sorry I missed that detail. Starting point is $2.5M. I like to use that because 4% of $2.5M is a nice even $100K.

Remember, any methodology that uses a fixed % of the remaining portfolio by definition has a 100% survival rate. If there is a multi-decade downturn, for example, your portfolio might be very small, but the same % of that portfolio would always be >$0 (though tiny).

 

[Dtail]

Sorry I missed that detail. Starting point is $2.5M. I like to use that because 4% of $2.5M is a nice even $100K.

Remember, any methodology that uses a fixed % of the remaining portfolio by definition has a 100% survival rate. If there is a multi-decade downturn, for example, your portfolio might be very small, but the same % of that portfolio would always be >$0 (though tiny).

Actually I see the 2.5mm now. Of course on 100% survival rate. Silly comment by me.
Thus the first 15 years of the retirement, the portfolio balance remains intact it appears, but at the detriment of having the withdrawals reduced by 50%. Not so great.
Is my assumption correct in that with VPW, due to higher starting WR% and the concept of NOT maintaining the original portfolio at the end, one would not have to take such drastic reductions in withdrawals?

 

[audreyh1]

Actually I see the 2.5mm now. Of course on 100% survival rate. Silly comment by me.
Thus the first 15 years of the retirement, the portfolio balance remains intact it appears, but at the detriment of having the withdrawals reduced by 50%. Not so great.
Is my assumption correct in that with VPW, due to higher starting WR% and the concept of NOT maintaining the original portfolio at the end, one would not have to take such drastic reductions in withdrawals?

Pretty much all these % of portfolio methods using ~4% are going to see ~55% reductions in income (withdrawals) and slow recoveries under the really bad sequence of returns scenarios. That’s just the way it is. If the portfolio gets cut in half or worse, your income will suffer proportionally.

Classic VPW still can see a large drop in income ~50%. It may start a little higher because initial withdrawal is usually higher than 4% depending on initial parameters. The recovery is faster because it’s drawing down the portfolio more aggressively.

 

[big-papa]

Pretty much all these % of portfolio methods using ~4% are going to see ~55% reductions in income (withdrawals) and slow recoveries under the really bad sequence of returns scenarios. That’s just the way it is. If the portfolio gets cut in half or worse, your income will suffer proportionally.

Classic VPW still can see a large drop in income ~50%. It may start a little higher because initial withdrawal is usually higher than 4% depending on initial parameters. The recovery is faster because it’s drawing down the portfolio more aggressively.

AudeyH1 is absolutely correct. The whole point of a variable withdrawal method is to reduce withdrawals when times are bad in order to preserve the portfolio. Classic VPW does likewise. And remember, classic VPW has an annual withdrawal rate that increases monotonically each year. In the early years, the % increases aren't all that big, but as time progresses, the increases becomes larger. In the 1966 starting case, you definitely see the effects of the portfolio losses dominating, but by the time you get to the go-go 80's you have a both high returns and larger withdrawal percentages as well.

What can you do about it? I've discussed using current valuations on this board as a modification to VPW. (The early-retirement now author does something similar as well in the posts above) If you don't like that, then another knob you have to turn is your AA. You could go more conservative with fewer stocks, for example. Again, in the late 60's/early 70's you had high inflation, so I would expect that TIPs, if they had existed then, would have pulled you out of the hole more quickly if you had replaced some of your bond holdings with it. And of course there are the "all-weather" type of portfolios such as Permanent Portfolio, Golden Butterfly, and Dalio's "All weather" which attempt to take into account most possible economic conditions - with the usual issue being that there just isn't enough historical data to know for sure for some of the holdings like commodities, Gold, or TIPs. And another issue being that these sorts of portfolios haven't historically out-performed during good times, so there is a psychological aspect to that as well, often referred to as tracking error.

Even with classic VPW, you can leave a legacy. The way it was written guarantees depletion in the final year. But you can add buffer years to it that would nearly guarantee something would be left over, you just can't know exactly how much since you don't know the date of your demise and you won't know how much is in the portfolio on that date ahead of time. You could re-write VPW using the PMT function in Excel, which has the capability of a fixed, nonzero amount remaining on the last day.

And you can go back to the Clyatt example I showed and up the percentage from 95% to 98+%. Just know that as you approach 100%, it starts to look more and more like SWR.

Or you can use SWR which just ignores what you're portfolio is doing and has a risk of premature depletion of your portfolio or a risk of having a huge amount left over. If you're happy living well below your means and high probability of leaving a legacy, then why not?

 

[Dtail]

Thanks @audreyh1/big-papa for your responses.
As of now, I will use the Clyatt concept keeping the buffer at 5%. However, I will use a 3%WR instead of 4% from 59 until 62 years old starting in 2019.
Once I hit 62, it gets much more complex for us with taking SS/pay down TIRA/convert Roth/managing ACA decisions all in play.

 

[Rianne]

I considered us FIRE because megacorp is out of our life and DH enjoys his consulting LLC and will continue to earn approx $42K/year for the next 3-4 years or so. He has complete flexibility in his life and I am fully RE.

We have a portfolio and a nestegg. I consider the portfolio totally separate than our nestegg. We are living off our nestegg and will until we're 65. The nestegg does not include DH consulting. Our portfolio is the total investment side of the picture and when I use VG nestegg calculations and FIREcalc calculations, I only use portfolio information. The reason for this is we only have I bond/EE bond and cash in nestegg with the offset of DH consulting. Those bonds earn approx 4%. Bought I bonds many years ago, same with EE bonds as a cushion when we were actively working. We have not used any of the bonds yet. So, the SWR doesn't mean a whole lot at this point in time as the future and our portfolio is entirely up in the air as we have learned over the last few weeks.