Confused when to rebalance

[belhawk]

Hi,

I'm a new bee to PP and started investing in this strategy recently. I've a question regarding when to rebalance.

Based on 15/35 band do rebalance only when an individual asset value hits this band or is it total asset value if it reaches 15/35 band?

Example, if I invested $40000
IWV - 10000
TLT - 10000
SHV - 10000
GLD - 10000

then do we rebalance when the asset value, example (IWV) reaches 16000 which is 35% of the total asset or
do we rebalance when the total asset value reaches 44000?

Thanks in advance !

[Xan]

Welcome, belhawk!

A rebalance is triggered by any one of your four components hitting 35% or 15% of the total portfolio's value.  Then you set them all back to 25%.

[belhawk]

Thx Xan!

So, I can have a situation where my total portfolio is up by 50% but couldn't rebalance as none have the assets have hit 15/35 band?

For example; if I continue with my same 40k investment and the market value is now:

IWV - 20000 - 33.2%
TLT - 15500 - 25.74%
GLD - 14700 - 24.41%
SHV - 10000 - 16.61%
Total - 60200

[Xan]

I wouldn't say it's ever necessarily wrong to rebalance, though.  15/35 isn't set in stone.  You can use other bands, or even just rebalance when you feel like it, as long as you are aware that there's a transaction cost, and that you might be leaving money on the table (but probably not much, and you might be doing better too).

As far as rebalance timings, as long as you stay within 15/35, I believe there's no really wrong answer.  Others may disagree.  But if rebalancing would help you sleep, then do it.

[belhawk]

Great! Thanks for the response.

[kobe1]

Welcome Belhawk.

Rebalancing also looks different when you are adding regularly to your portfolio.  Some people recommend adding all new money to cash and then rebalancing when cash gets to be 35% of your total PP.  Other people buy lagging assets when adding new money.  Thereby restoring a 4 x 25 portfolio once or twice a year.  As others have mentioned, the PP is quite simple in concept, but full of nuances in its implementation.

Good Luck!

[nikao]

I see a lot of backtests on results being done and reported on this forum, but is there also a backtest of results available on rebalancing in different ways?
Using different bands, or frequencies?
Would find that quite interesting to see what the impact would be on results

[steve]

I start to watch my portfolio when I reach the 20/30 bands, as it gets closer to the 15/35 bands I look for opportunities like tax loss harvesting etc. I usually re-balance as it gets closer to the 15/35 but before it reaches it.  I enjoy re-balancing because then I can relax and go a long time without looking. I just re balanced this Dec 4 2015 the re balance before was June 26 2013 two and a half years ago. Just a little side note, I may have waited a little longer but with the Sprott GTU takeover battle I thought it was time to take tax loss harvest with GTU add to my physical core bullion holdings and have a smaller exposure to online gold.

[Jack Jones]

Rebalancing also looks different when you are adding regularly to your portfolio.  Some people recommend adding all new money to cash and then rebalancing when cash gets to be 35% of your total PP.  Other people buy lagging assets when adding new money.  Thereby restoring a 4 x 25 portfolio once or twice a year.  As others have mentioned, the PP is quite simple in concept, but full of nuances in its implementation.

You can also split your contributions, and contribute a fourth to each asset. Then, just rebalance as normal.

[nikao]

I see a lot of backtests on results being done and reported on this forum, but is there also a backtest of results available on rebalancing in different ways?
Using different bands, or frequencies?
Would find that quite interesting to see what the impact would be on results

I looked at the effect of rebalancing band width on a 6-year data set a few years ago.  My study showed that 15/35 bands were best.  However, what I would say now is that I suspect that details matter a lot.  For example, in my study, I used daily data. Furthermore, I assumed that the allocation was sampled at the end of each day and, if bands were violated, the rebalance was implemented prior to the start of the next trading day.  But I can think of a bunch of wrinkles on this.  For example:

* What if you check balances monthly?
* What if you check balances at no regular intervals?
* What if there is a delay (systematic or random) between the rebalance trigger and the actual rebalancing?
* What if, during the delay, the money sits in "cash"?

etc.

I would not recommend using backtesting as a means to optimize rebalance band width.

Thx, that's exactly what I was looking for. I agree with your remarks, but still nice to have some guidance by actual numbers as to whether 5% or 10% bands are better, apart from the practical point of view

[Xan]

Rebalancing also looks different when you are adding regularly to your portfolio.  Some people recommend adding all new money to cash and then rebalancing when cash gets to be 35% of your total PP.  Other people buy lagging assets when adding new money.  Thereby restoring a 4 x 25 portfolio once or twice a year.  As others have mentioned, the PP is quite simple in concept, but full of nuances in its implementation.

You can also split your contributions, and contribute a fourth to each asset. Then, just rebalance as normal.

Another option is to contribute proportionally, that is, if you're at 30/20/25/25, then contribute in the same proportions.  This is the option that prevents contributions from affecting the timing of your rebalances.

[Cortopassi]

Welcome, belhawk!

So, I can have a situation where my total portfolio is up by 50% but couldn't rebalance as none have the assets have hit 15/35 band?

Yes, indeed.

For example; if I continue with my same 40k investment and the market value is now:

IWV - 20000 - 33.2%
TLT - 15500 - 25.74%
GLD - 14700 - 24.41%
SHV - 10000 - 16.61%
Total - 60200

As you surmised, you would not rebalance in this situation, because all of the asset allocations fall within the 15/35 limits.

Yes, that will be a grand day, when I can see my PP up 50%! 

[dualstow]

Another option is to contribute proportionally, that is, if you're at 30/20/25/25, then contribute in the same proportions.  This is the option that prevents contributions from affecting the timing of your rebalances.

I've never seen or thought of that before. What an interesting idea.

[Jack Jones]

Rebalancing also looks different when you are adding regularly to your portfolio.  Some people recommend adding all new money to cash and then rebalancing when cash gets to be 35% of your total PP.  Other people buy lagging assets when adding new money.  Thereby restoring a 4 x 25 portfolio once or twice a year.  As others have mentioned, the PP is quite simple in concept, but full of nuances in its implementation.

You can also split your contributions, and contribute a fourth to each asset. Then, just rebalance as normal.

Another option is to contribute proportionally, that is, if you're at 30/20/25/25, then contribute in the same proportions.  This is the option that prevents contributions from affecting the timing of your rebalances.

Ah, that's the aim I was trying to achieve, but my methods were off. Thanks!

[nikao]

Another option is to contribute proportionally, that is, if you're at 30/20/25/25, then contribute in the same proportions.  This is the option that prevents contributions from affecting the timing of your rebalances.

I've never seen or thought of that before. What an interesting idea.

I don't get it I think. Why contribute in the same proportions instead of the other way around to get back into the desired 25/25/25/25? (so contribute 20/30/25/25 if you are currently at 30/20/25/25)?

[dualstow]

I don't get it I think. Why contribute in the same proportions instead of the other way around to get back into the desired 25/25/25/25? (so contribute 20/30/25/25 if you are currently at 30/20/25/25)?

I don't think there's anything wrong with contributing inverted proportions as you're suggesting. It's like buying the lagging asset, i.e. buying gold when that is the asset that's down. Sophie checked all this out once, though, and it turns out that buying the lagging asset was statistically near the bottom of a list of strategies' performances. If I remember correctly.

Contributing in the same proportions is perhaps a bit like momentum investing. If stocks are on a roll, why skip contributing more to stocks? Because stocks are "expensive"? Well, that can only be known in the future. And, if you're still contributing to the other 3 assets, in proportion, you're not changing much.

I think the takeaway is to pay more attention to proportion than to current performance of a specific asset. Current performance is an allusion, albeit a powerful one. Ignore it and pay attention to bands & allocations.
----
Btw, some people just rebalance once a year to get things back on track, especially if no bands were hit. That's ok, too.

[sophie]

Details in an old thread, but cliff notes version:  I ran backtests for three strategies:  1) contribute to cash, 2) contribute to lagging asset, and 3) contribute to all 4 assets in equal proportions. 

Contributing to cash had the worst performance, don't remember exact numbers but it was quite significantly worse than option #3, which had the best performance.  Contributing to lagging asset was not too far behind.  This however ignores tax consequences of rebalancing, which can be quite significant.

So for a taxable PP and a high marginal tax rate, you want to contribute to the lagging asset.  Otherwise, your best bet is simply to divide contributions among all assets equally, or do with Pointed Stick does and alternate assets purchased each month to minimize transaction fees.

I didn't run a test of contributing to maintain existing percentages, but that struck me as a method that would probably be way too difficult to sustain over time.  If anyone wants to run those simulations feel free!

[nikao]

That's all great info, thx!
One of the important things for me is to automate everything as much as possible and keeping it simple. That way I can make sure it'll be a true passive portfolio in the sense that I'm not tempted to time the market or anything
So mostly fixed monthly (automated) contributions works best for me, while paying attention to the bands ..