SWR of 4%+COLA or reset clock to 4%?

[pb4uski]

...Note that it does reduce the average ending portfolio value, which is probably sort of what you want ...

Yes, realistically, I'm more concerned with dying with too much money than not enough. But that said, we could afford to spend more now but don't... in part because DW is so [-]cheap[/-] frugal.

 

[garyt]

Resetting to a new 4% feels good when your portfolio is bigger than when you first set it. But you are also resetting your Sequence of Returns Risk (SORR).

This would be my concern. By already having 3 good years in, you're pretty much golden regarding SORR. But if you reset, you're back in danger of a bad year in your first few years again. But the 4% rule already has SORR built in so there's that.

 

[RunningBum]

But the 4% rule already has SORR built in so there's that.

Isn't the 4% rule generally thought to have a 95% rate? So when you say it "has SORR built in", isn't an SORR situation more likely to be one of the 5% cases? So yeah, it's built in... as a very possible failure case.

 

[tb001]

Isn't the 4% rule generally thought to have a 95% rate? So when you say it "has SORR built in", isn't an SORR situation more likely to be one of the 5% cases? So yeah, it's built in... as a very possible failure case.

Yes, this is how I think of it as well. As others have said, you would be increasing the SORR. That said, I think it’s actually 4.5% that gets you to 95% success, assuming a 30yr retirement.

My hope is that we are on a good enough trajectory that our portfolio grows enough to turn our initial higher withdrawal rate into 3-3.5%. At that point, I’d be comfortable adjusting up with further gains (I.e. adjusting up to keep it at 3-3.5%).

 

[ERD50]

Isn't the 4% rule generally thought to have a 95% rate? So when you say it "has SORR built in", isn't an SORR situation more likely to be one of the 5% cases? So yeah, it's built in... as a very possible failure case.

That's why I like to demonstrate the "retire again and again" approach using a historically 100% safe initial WR% (something ~ 3.3% for 30-40 years). Sticking strictly to the historical data, if you are 100% safe for that period, then you can ratchet the spending up if the portfolio grows faster than inflation after withdrawals, and you still never fail. Since that rate passed in all cases in history, you can apply (re-apply) it in all cases in history. The market doesn't know when you retired - you are just another point on the line!

Now obviously, taking out more money will reduce your buffer when you approach those years that would have failed if you took out more, but you still don't fail. But, within that data set, it is the absolutely most efficient approach to pulling the most out w/o hitting failure. The range of outcomes is reduced (less to heirs on average), because you made the best use of withdrawals over that history. You stay < $0, and minimize the excess. It's just arithmetic within that data set, there's no better way to do it if you are tossing the dice and could land on any particular starting year. As an example, if you did land on that worst case year, where the "SWR" left you right at zero $ in your final year, the "retire again and again" approach would never have a year where you increase your withdrawal. That's just the way it works. Follow the squiggly lines and you'll see.

-ERD50

 

[RunningBum]

That's why I like to demonstrate the "retire again and again" approach using a historically 100% safe initial WR% (something ~ 3.3% for 30-40 years). Sticking strictly to the historical data, if you are 100% safe for that period, then you can ratchet the spending up if the portfolio grows faster than inflation after withdrawals, and you still never fail.

I'll buy that. If it turns out to be a worse case than any of that historical data I'd assume there's fat to cut.

 

[Navigator]

This is an interesting question. The 4% rule (updated to 4.5%) is based on historical data and is the very lowest withdrawal rate that would have been absolutely safe, using historical returns. Most of the retirement periods since 1928 have had much higher SWRs. Unless one wants to leave their heirs a small fortune, some sort of adjustment is warranted. I would like to point out two things, though.

1) If you “reset” your withdrawal rate in the manner you describe, you will lose all information on what the market has done since you retired. If one believes the market regresses to its mean P/E ratio, then if the market did well in the last few years it will likely not do so well in the next few. This information will be lost.

2) Stocks have fairly low average returns for the amount of volatility in prices. In engineering we call this a small signal to noise ratio. Moreover, there are really only 3 independent 30 year periods since 1928. For example, someone retiring in 1986 will have 29/30 years in common with someone retiring in 1987, and these are not independent of each other. All this makes me think we should not have a lot of confidence in the 4% rule.

All this is a long-winded way of saying I think your emphasis on flexible withdrawal rates is good, but I would be cautious about withdrawing too much money simply because the stock market is up, especially in the first several years of retiring. Of course, I have been known to be wrong before.

 

[Rambler]

We havent followed the “rule” as hard and fast. I don’t think we’ve ever gone beyond 4%, and rather than use 4% plus inflation, we just look at recent returns and tighten a bit or loosen a bit. We did tighten quite a bit in March this year when the markets were slammed, but then again, was it intentional? Or was it a result of lost spending opportunities? In a normal year, we’t take probably 8-10 5 day trips plus a couple of two-ish week trips. This year, we’ve had one weekend trip and one five day trip...because of COVID. But, i always had those massive unrecognized losses in my mind, and both of us have been just a bit more mindful...not seriously trying to save money, just mindfulness. Now that we are just about back to normal, you’d think we’d be back to our regular spending patterns, but you’ll be wrong. The big ticket items havent changed much (phoenix summer electric bills) but the discretionaries are down significantly.

I suspect that for most years, we’ll end up around 3% WR, but we’ve been up to around 4% and will be around 1.8-2.0% this year...no inflation adjustments, jut being mindful of the comings and goings, and keeping them lower than what we earn.

 

[meleana]

I don’t use withdrawal percentages per our Financial Advisor. I withdraw what we need allowing for some unexpected overages in our budget.

 

[audreyh1]

Kitces had a well thought out paper that he called “Ratcheting Withdrawals” with some realistic formulas. The idea is to not continuously grow your portfolio to far larger than it was (inflation adjusted). It works both ways, of course, and was, IMHO, a good solution as long as you don’t have to have your entire income from a steadily rising COLA from your portfolio. For those fortunate enough that the portfolio only provides discretionary income, that works great, you get to spend more when you make more and less if you lose money. As pb4uski mentioned, there is of course a larger chance of failure because you are spending more (duh), but since the chances are greater that you will increase your portfolio bs lose it all, it really depends on your risk tolerance.

Right - Kitces did have a very thoughtful approach to this. But I thought that income did go up, never down. He just had more complex criteria for when it was OK to ratchet up.

 

[Koolau]

I don't need to justify spending more money, but would be interested in the "retire again and again" process if I could have the whole 30 years again.

I have already spent 8 of those.

Mentioned elsewhere, I'm 15 years through my 30 years. (Dang, where does the time go?) So, now I've reset to age 100 which means I have 26 years left. I'll believe it if you will. I guess that means if I DO make 100, I'll have to use the 9mm (final) solution. On the bright side, I'm not consistently any place close to 4% (and especially not 4.5%). I am still concerned about future inflation, but the only inflation right now is in my portfolio - meaning I can spend more, most years. But I think that's cheating.

Anyway, I think as long as you correctly fill in the numbers in FIRECalc, you can re-retire every year if you want to. Just be certain you are happy with the % survival rate that comes out the back. YMMV

 

[GravitySucks]

I did a reset at year 5 (access to IRA), and plan to reset again at year 10 (access to Medicare), and again at year 15 (SS max age).
Yrs 1-5, 4.5% of initial balance.
Yrs 5-10, 5% of balance as of 2018, almost 6% initial balance.

Now if I could only bring myself to actually spend it.