Does re-allocation require selling in order to be most effective?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
bironology
Posts: 146
Joined: Sun Nov 08, 2015 6:18 pm

Does re-allocation require selling in order to be most effective?

Post by bironology »

Does re-allocation of a portfolio require selling of overweight equities to maximize the benefit?

I'm at a high income period of my life, so each bi-weekly period brings new money into both tax-advantaged and taxable accounts. If I review my AA quarterly (and I do, sometimes more often) and a category is getting out of band, I have simply been redirecting future purchases to nudge the AA back to where it should be. This requires some patience and tolerance for being +-2% here and there, but it all works out. I don't stress out that my US-LARGE allocation is 42.5% when I wanted it at 40%, I just modulate the future investments to some other category that is too skinny. I think many of us do this.

When I think about the concept of re-allocation as a means to capitalize on market swings without explicitly market timing, I wonder if I'm only taking advantage of one half of the phenomenon. If one category is riding high and I don't sell off some of it, I never "take profit". Conversely, if one category just had the bottom fall out of it, with my current scheme of modulating future purchases, it could take 6 months or more for me to fully capitalize on the buying opportunity. I've been more timid about selling in my taxable account because I hate paying taxes, but maybe I'm optimizing the wrong metric.

I could imagine building a mathematical model of this, and I almost did it just now, but figured I should be lazy instead and see if someone else has covered this topic somewhere.
User avatar
AtlasShrugged?
Posts: 699
Joined: Wed Jul 15, 2015 6:08 pm

Re: Does re-allocation require selling in order to be most effective?

Post by AtlasShrugged? »

bironology....I do the same thing = redirect future contributions.

If transaction cost is not a factor, the math (from what I have read) says that there is a small benefit to selling. Some time ago, there was a thread on selling when you had a positive variance. Livesoft was all over that thread. If you find that thread, you'll see the lively debate.

Net net: A small benefit. My understanding is that we reallocate for risk, not necessarily to lock in gains.
“If you don't know, the thing to do is not to get scared, but to learn.”
Topic Author
bironology
Posts: 146
Joined: Sun Nov 08, 2015 6:18 pm

Re: Does re-allocation require selling in order to be most effective?

Post by bironology »

User avatar
AtlasShrugged?
Posts: 699
Joined: Wed Jul 15, 2015 6:08 pm

Re: Does re-allocation require selling in order to be most effective?

Post by AtlasShrugged? »

No, that ain't it.

I have a vague recollection that the discussion centered around a claim (ultimately shown to be wrong) of a 3% to 4% rebalancing 'bonus'. As I recall the thread, weekly monitoring of your portfolio coupled with rebalancing whenever the positive variance went above 3%-4% in your portfolio yields a benefit over the long haul (for 60/40, 80/20 portfolios).

The 'math' from what I recall was the benefit was more along the lines of 0.2% to 0.3%, and at that, only if you were super consistent and almost never missed a rebalance. My takeaway: Yeah, there is a very, very slight benefit, but you really need to work at it to get it. Not sure it would be worth the effort unless it was a completely automated process. And I don't have a completely automated process. :(

I know Livesoft, Simplegift and longivest were all over that thread. I learned a lot from that thread (just by researching what these guys were writing). I distinctly recall it was pretty lively.
“If you don't know, the thing to do is not to get scared, but to learn.”
livesoft
Posts: 75145
Joined: Thu Mar 01, 2007 8:00 pm

Re: Does re-allocation require selling in order to be most effective?

Post by livesoft »

You are problably thinking of the Daryanani paper on opportunistic rebalancing:
http://resource.fpanet.org/resource/09B ... yanani.pdf

I'll just say in this thread that if you want higher return, you will probably need a higher percentage to equities on average. Delaying rebalancing when equities go up can get that.*

I'm not sure what the OP means by re-allocation? Is that shifting from bonds to equities when an equity asset class is not keeping up with other equity asset classes in a rising market?

*Maybe. The 10-year average annual returns reported by morningstar.com for the 60/40 LifeStrategy Moderate Growth and the 80/20 LifeStrategy Growth funds differ by only 0.48% when I looked just now. That's almost like saying that if you can save 0.5% per year on your investment fees, expense ratios, commissions, taxes, and other costs, that you get the return of a more risky portfolio without the risk. Or conversely, if you paid an advisor a mere 0.5% of AUM each year to invest in the Vanguard LifeStrategy Growth fund for the past 10 years, then you would only be left the return of the Vanguard LifeStrategy Moderate Growth fund after many years.
Wiki This signature message sponsored by sscritic: Learn to fish.
mega317
Posts: 4582
Joined: Tue Apr 19, 2016 10:55 am

Re: Does re-allocation require selling in order to be most effective?

Post by mega317 »

If one asset class was outperforming relatively and you sold, then your new contributions would go more towards that class than if they were otherwise solely dedicated to the underperforming class. If you're staying within 2% that is plenty tight, more than most people do. If there is a major stock crash then I probably wouldn't wait 6 months to rebalance and would do some selling then.
https://www.bogleheads.org/forum/viewtopic.php?t=6212
Topic Author
bironology
Posts: 146
Joined: Sun Nov 08, 2015 6:18 pm

Re: Does re-allocation require selling in order to be most effective?

Post by bironology »

To clarify my question about re-allocation, I suppose the term should have been rebalancing to maintain my target allocation. I don't really distinguish between whether this is across equity/debt categories. My target AA is

30% US large cap
20% US small and mid cap
20% international
30% bonds

My question was really when to trigger a sale of one type of investment in order to correct an imbalance. I think the key is how long it would take to get into balance. 2% deviation is easily rectified with new money. But a market drop in one of those AA categories may create an imbalance that would take months to correct with new money alone, so a sale and buy would be needed to correct that without waiting a year.
User avatar
WoodSpinner
Posts: 1903
Joined: Mon Feb 27, 2017 1:15 pm

Re: Does re-allocation require selling in order to be most effective?

Post by WoodSpinner »

FoP,

Many of us use a set of bands to trigger rebalancing so we don’t focus on smaller imbalances.

For instance in my IPS, I use:
  • > 5% out of balance based on portfolio allocations (e.g. trigger on a 30% balance if target is 25%)
  • > 25% out of balance on an individual holdings ( e.g. trigger on a 7.5% balance if target is 5%)
I wrote a little Utility to help with the analysis and coordinate the process.
viewtopic.php?t=221267

For smaller imbalances redirecting future investments is a great idea.

Anyway, that is how I handle things...

Good luck

WoodSpinner :)
livesoft
Posts: 75145
Joined: Thu Mar 01, 2007 8:00 pm

Re: Does re-allocation require selling in order to be most effective?

Post by livesoft »

Let me ask a specific question then: Did you rebalance in to small/mid cap in July, August, or September?

Here's the cool thing: It didn't matter if you did. :)

On 8/18/2017, morningstar.com shows that VSIAX had a 2017 total return of about -3%. That is, the best day to buy it in 2017. But look what has happened since that date:

Image

All my favorite equity funds have gone up 13% to 14% just like the SCV (VSIAX). That is, if you bought VSIAX at its low for 2017 by exchanging money from another equity fund, you didn't help your portfolio performance one bit.

You could not have timed the purchase of VSIAX better in 2017 and it didn't matter -- as long as you didn't own it before 8/18.
My question was really when to trigger a sale of one type of investment in order to correct an imbalance.
My advice is to follow the RBD strategy, but that may be too much market timing for you. See the red line in the chart for DGS? DGS had an RBD in late September.

Edit to add: Of course, one should use dividends / distributions for rebalancing, too. This should keep one's portfolio mostly in balance anyways. For instance, in late September there were all those quarterly dividends that could be used to buy DGS or another small-cap value investment if desired.
Last edited by livesoft on Fri Jan 05, 2018 10:51 am, edited 1 time in total.
Wiki This signature message sponsored by sscritic: Learn to fish.
marklar13
Posts: 123
Joined: Wed Jul 08, 2015 11:47 am

Re: Does re-allocation require selling in order to be most effective?

Post by marklar13 »

I think you should be content to just re-balance through new contributions if that will cover it. If new contributions alone can't get you to within the AA that you want then exchange funds as needed. Unless your re-balancing thresholds are very tight, you will very rarely need to sell. It takes a really sizable run-up to trigger a 5% re-balancing band (which I believe is a common threshold on this board). For example, if stocks return 20% and bonds return 0%, then this would cause your 70/30 allocation to shift only about 3.7% (I realize this ignore the equity splits... just an example).

I'm a believer that re-balancing between equities and fixed income is about managing risk, not going for more return. In fact, I think if you are chasing higher return then you are very likely better off letting your equity allocation drift higher when they rise (or re-balancing "aggressively" into equities when they fall). Equities have a higher expected return than fixed income... so the more you hold the higher the expected return of your portfolio will be.

It probably gets a little more complicated than that within the equity allocation, since comparing expected return of US, Intl, Small cap, etc is a little fuzzier. But... as long as you are sending new contributions to the asset class that is the furthest below its target, then you are essentially "buying low" in theory (compared to the other asset classes).
TwstdSista
Posts: 996
Joined: Thu Nov 16, 2017 4:03 am

Re: Does re-allocation require selling in order to be most effective?

Post by TwstdSista »

OMGoodness thank you for this thread. I've been implementing my IPS (after a few mis-steps) and realized I'd be low on TIM. I intend to fund our 2017 Roths prior to April, but those don't hold TIM. What to do, what to do. Doh -- my i401k! I had not yet decided what to put in there once I fund it. TIM it is!

So thank you. (although trading some TSM for TIM in a Roth isn't a bad idea either....)
Topic Author
bironology
Posts: 146
Joined: Sun Nov 08, 2015 6:18 pm

Re: Does re-allocation require selling in order to be most effective?

Post by bironology »

livesoft wrote: Fri Jan 05, 2018 10:36 am Let me ask a specific question then: Did you rebalance in to small/mid cap in July, August, or September?
I had a windfall of cash in that timeframe so I brought my allocation into balance by buying. So I did rebalance but did not need to sell anything to do so. If I had that need, I would have pondered it very carefully.
...
livesoft wrote: Fri Jan 05, 2018 10:36 am Edit to add: Of course, one should use dividends / distributions for rebalancing, too. This should keep one's portfolio mostly in balance anyways. For instance, in late September there were all those quarterly dividends that could be used to buy DGS or another small-cap value investment if desired.
Yes and that's exactly what I did. I do not use automatic dividend reinvestment because I'd rather make tactical choices about how to reinvest based on my AA. I'm also concerned about having many small lots but I know that's not a big deal with today's brokerages - still I like to be in the driver's seat.

And for the record, I'm not interested in market timing unless there's a once in a lifetime cataclysmic event - if such an event exists. I'm very closely aligned to BH principles (but I'm no disciple! :happy )
Topic Author
bironology
Posts: 146
Joined: Sun Nov 08, 2015 6:18 pm

Re: Does re-allocation require selling in order to be most effective?

Post by bironology »

marklar13 wrote: Fri Jan 05, 2018 10:48 am I think you should be content to just re-balance through new contributions if that will cover it. If new contributions alone can't get you to within the AA that you want then exchange funds as needed. Unless your re-balancing thresholds are very tight, you will very rarely need to sell. It takes a really sizable run-up to trigger a 5% re-balancing band (which I believe is a common threshold on this board). For example, if stocks return 20% and bonds return 0%, then this would cause your 70/30 allocation to shift only about 3.7% (I realize this ignore the equity splits... just an example).
OK, my takeaway is that I need to add rebalancing bands into my IPS. That's not a concept that I've had any rigor around, thus have had no formula. I should lock that down. Will begin research...
Post Reply