I should rebalance but..................

[big-papa]

Similar to others.
Taxable accounts: 100% stocks
Deferred accounts: bonds + stocks

Across the two types of accounts, I can do all rebalancing totally within the deferred accounts without having to touch the taxable accounts. But I still contribute to my taxable accounts and direct funds towards whichever stock asset sub-type is below target. Likewise, I direct cap gains/dividends to my cash position and then I direct the funds to whichever assets are out of whack. If the addition of such funds in the taxable accounts pushes an asset sub-type too high, then I'll rebalance out of something similar in the deferred accounts.

And among the methods I showed earlier, I use the 5% absolute for major asset classes and 25% relative for sub-classes. I only rebalance when these are exceeded with no thought to the calendar.



When I do rebalance, I don't always rebalance all assets - just the ones out of whack. And I don't always rebalance all the way to the target allocation.


Someday I will be RE - at that time, much of the rebalancing can be done by choosing where the withdrawals come from..

 

[big-papa]

That column compares rebalancing annually vs. not at all, so it makes no case that annual is better than rebalancing more often. That said, I only rebalance once a year, but if I have money to add (from a CD maturing, perhaps) or withdrawing money, I will add to the asset class that is low, or take from the one that's high, so I do some ad hoc rebalancing at other times.

Actually the article itself discusses the 5/25 rule from Larry Swedroe which is neither annual rebalancing nor no rebalancing. It's only the graphs that show annual rebalancing vs. no rebalancing. Weird that the author chose not to include 5/25 in the graph since that was the primary subject of the article.

Anyway, something to be aware of when trying to backtest any sort of rebalancing method that isn't based purely on rebalancing on a certain date. That is, the % held in each asset type on your starting date for the backtest matters as well as the actual starting date itself. The "initial condition" for you engineer/math types changes the final answer to a greater or lesser degree depending on the exact method used to rebalance.

 

[jazz4cash]

I should rebalance but just because you have an ISP doesn’t make it easy to buy equities in this market. My policy state annual rebalance near the beginning of the year. I’ve also managed to box myself in by having CDs in lieu of bonds. I trust I can sort this out along with next years withdrawal to stay in the ballpark of 60/40. I’m very tolerant of a ballpark vs a precise AA target.

 

[Dtail]

Actually the article itself discusses the 5/25 rule from Larry Swedroe which is neither annual rebalancing nor no rebalancing. It's only the graphs that show annual rebalancing vs. no rebalancing. Weird that the author chose not to include 5/25 in the graph since that was the primary subject of the article.

Anyway, something to be aware of when trying to backtest any sort of rebalancing method that isn't based purely on rebalancing on a certain date. That is, the % held in each asset type on your starting date for the backtest matters as well as the actual starting date itself. The "initial condition" for you engineer/math types changes the final answer to a greater or lesser degree depending on the exact method used to rebalance.

Question on the 5/25 rule.
If for example one has 4 sub categories of stocks and one moves down 25%, but the others move down also but less than 25%, wouldn't the natural weighting of the 4 classes against each other end up causing no category to be down 25%?
Am I making sense?

 

[big-papa]

Question on the 5/25 rule.
If for example one has 4 sub categories of stocks and one moves down 25%, but the others move down also but less than 25%, wouldn't the natural weighting of the 4 classes against each other end up causing no category to be down 25%?
Am I making sense?

Yes, that can be the case. But it's not about how much a particular asset moves by itself. It's always relative to the total amount in your portfolio. If you have 50% stock and 50% bond and both drop by 25%, then you still have 50% stock and 50% bond, so you don't rebalance.

The 5/25 rule works like this.
For Major asset classes (like stocks), suppose you have 50% as your target. You would rebalance if it is >55% or <45%.

But you might also do this to one of the individual pieces as well if it's a larger asset. Suppose that stock is 40% TSM and 10% Small Caps. You would rebalance if TSM >45% or <35%. The Small Caps, being a smaller target allocation of only 10% would fall under the 25% rule. That is if it is <10*0.75=7.5% or >10*1.25%=12.5% then you would rebalance it.

Here's a decent explanation of it:
https://www.whitecoatinvestor.com/rebalancing-the-525-rule/

 

[Dtail]

Yes, that can be the case. But it's not about how much a particular asset moves by itself. It's always relative to the total amount in your portfolio. If you have 50% stock and 50% bond and both drop by 25%, then you still have 50% stock and 50% bond, so you don't rebalance.

The 5/25 rule works like this.
For Major asset classes (like stocks), suppose you have 50% as your target. You would rebalance if it is >55% or <45%.

But you might also do this to one of the individual pieces as well if it's a larger asset. Suppose that stock is 40% TSM and 10% Small Caps. You would rebalance if TSM >45% or <35%. The Small Caps, being a smaller target allocation of only 10% would fall under the 25% rule. That is if it is <10*0.75=7.5% or >10*1.25%=12.5% then you would rebalance it.

Here's a decent explanation of it:
https://www.whitecoatinvestor.com/rebalancing-the-525-rule/

Thanks. That is the way I understood it.

 

[pb4uski]

I rebalance within my traditional IRA. It is large enough relative to my other accounts such that I should never have a problem achieving my desired AA entirely within that account. Therefore, no (immediate) tax consequences to rebalancing.

I believe there are others who do this as well.

+1

+2 I think this would be the case for anyone whose tax-deferred balances exceed their fixed income allocation... which is true in our case.

 

[steelyman]

The 5/25 rule is new to me, interesting reading in those links. I guess I’m a 5-only rebalancer regardless of the target percent.

When I do rebalance if necessary in January, it takes me a few days to reshuffle things for the upcoming year. I sell/buy, check, sell/buy again (with different asset classes), until done.

Is that usual or are others able to do it in one batch of trades on a single day?

 

[USGrant1962]

Another thing to keep in mind is that the math on rebalance bands is not a straightforward as you might think. I'll give two examples:

1. Say you are 50/50 AA with a +-5 rebalance band. One would think that a 10% drop in equities would hit your band (10% of 50% is 5%). Not so. It turns out that, everything else constant, you are only about half way to your band limit.

In a $100K portfolio, stocks would have dropped from $50K to $45K. But, assuming FI is constant, your portfolio also dropped to $95K. 45K/95K = 47.4%. You are only 2.6% into your 5% rebalance band.

2. The other issue is that when using % bands, gains and losses are not reciprocal. The classic example is a 50% loss. To get back to your starting portfolio you need a 100% gain.

Taking these two into account, I've personally settled on +6% / -5.5% rebalance band (around a nominal 60/40 AA).

Kitces has a good discussion here https://www.kitces.com/blog/best-op...y-time-horizons-vs-tolerance-band-thresholds/

 

[big-papa]

The 5/25 rule is new to me, interesting reading in those links. I guess I’m a 5-only rebalancer regardless of the target percent.

When I do rebalance if necessary in January, it takes me a few days to reshuffle things for the upcoming year. I sell/buy, check, sell/buy again (with different asset classes), until done.

Is that usual or are others able to do it in one batch of trades on a single day?

5-only works very well if you only have a few asset classes - for example if you use a 2 or 3-fund type of portfolio (TSM/Int'l/bonds) then something like 5/25 doesn't do anything for you as long as the 2 or 3 funds have a reasonably high percentage target for each.

In the past, when I did it only annually and did a full rebalance of everything back to targets it would take me a few days. But with 5/25, typically I'm only rebalancing one or two assets and I can usually get it done in a day - sometimes two..

 

[steelyman]

5-only works very well if you only have a few asset classes - for example if you use a 2 or 3-fund type of portfolio (TSM/Int'l/bonds) then something like 5/25 doesn't do anything for you as long as the 2 or 3 funds have a reasonably high percentage target for each.

In the past, when I did it only annually and did a full rebalance of everything back to targets it would take me a few days. But with 5/25, typically I'm only rebalancing one or two assets and I can usually get it done in a day - sometimes two..

Interesting, thanks for the comments. I’ll pay attention to my situation in January to see how it would be different. It may be for me as I’m currently using 4 classes for stocks, 4 for bonds, and one for real estate.

The smallest allocations are for emerging stocks/bonds and high-yield US bonds.

 

[big-papa]

Another thing to keep in mind is that the math on rebalance bands is not a straightforward as you might think. I'll give two examples:

1. Say you are 50/50 AA with a +-5 rebalance band. One would think that a 10% drop in equities would hit your band (10% of 50% is 5%). Not so. It turns out that, everything else constant, you are only about half way to your band limit.

In a $100K portfolio, stocks would have dropped from $50K to $45K. But, assuming FI is constant, your portfolio also dropped to $95K. 45K/95K = 47.4%. You are only 2.6% into your 5% rebalance band.

2. The other issue is that when using % bands, gains and losses are not reciprocal. The classic example is a 50% loss. To get back to your starting portfolio you need a 100% gain.

Taking these two into account, I've personally settled on +6% / -5.5% rebalance band (around a nominal 60/40 AA).

Kitces has a good discussion here https://www.kitces.com/blog/best-op...y-time-horizons-vs-tolerance-band-thresholds/

All true things. I've kept with 5/25 but I have a spreadsheet that monitors the % of everything and automatically flags when something is out of whack.

There was an interesting discussion on bogleheads about 5/25 a couple of years ago...
https://www.bogleheads.org/forum/viewtopic.php?f=10&t=185596

Regardless, as far as I can tell by having backtested just about every method I've read about, plus a few I came up with, I concluded:
1) there is something to be said for simplicity
2) No rebalance method I've run into will ever give you consistently higher returns - this is more about managing risk.
3) Anything you can do that keeps transaction costs and taxable events few and small is a good thing.


I have seen some differences by choosing wider bands, but that makes sense - you're letting stocks drift more and we already know that over a very long period of time, not rebalancing at all (or even holding a 100% stock portfolio) will eventually outperform the bond portion of a portfolio, as long as you can tolerate the volatility...

 

[corn18]

5/25 guy here. Enough room in 401k to do most of the rebalancing. Although towards the end of 2017, I was running out of room to buy bonds. The market stepped in and took care of that for me.

35% Total US Stock
25% Total International Stock
40% Total Bond

I need a T-shirt that says Boring Boglehead on the front and Don't just do something, Stand There! on the back.

 

[Christine]

One question on the 5/25 rule:


If you have the common 60/40 stocks/bonds. And then using the 25 rule you need to buy more of some index fund.


Do you then sell bonds to finance this or the best performing among your other stocks?


Another way of putting it - do you rebalance within stocks or across the whole portfolio?

 

[big-papa]

One question on the 5/25 rule:


If you have the common 60/40 stocks/bonds. And then using the 25 rule you need to buy more of some index fund.


Do you then sell bonds to finance this or the best performing among your other stocks?


Another way of putting it - do you rebalance within stocks or across the whole portfolio?

If you're using the 25 rule then you must have some funds in the stock and/or bond portions that are 20% or less of the total portfolio. Otherwise only the 5 rule would ever apply if you just have a single stock and single bond position.

If you do have more than 2 funds and one or more holdings is less than 20%, then those funds are subject to the 25 rule. If the rule is exceeded on the positive side for any fund then you sell some of it and buy any of the funds you have which are below their target allocations. You don't have to rebalance all the way back to the targets and you don't have to just purchase more of just one fund. You can spread the joy.

Vice versa if the 25 rule is exceeded on the negative side - sell some of any fund or funds that are above target and buy more of the one that was too low. The final result doesn't need to be that all funds are all the way back to their original target. Being back inside the the limits is fine.