Jason Hsu and William Bernstein Like Overbalancing

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Mon Sep 12, 2016 9:21 am

http://news.morningstar.com/cover/video ... ?id=768821

Hsu: "take money away from what's done really well and become expensive and then move money into what's done poorly...So if you do this, you know, a systematic rebalancing, or maybe do a little bit more, a systematic over-rebalancing, you could develop an edge certainly, over people who are more trend-chasing in how they make decisions"

Kinnel: "you're suggesting you could even go another step further and even add a little more to the worst-performing investments in your portfolio, take a little more away from the strongest performers"

Hsu: "That's absolutely true, and like you say, within a tax-advantaged account, the benefit of rebalancing is phenomenal because it doesn't have the tax consequence associated with it. I've even seen research, where careful and infrequent rebalance is still a lot better even in the face of some tax consequences."

William Bernstein, in "The Intelligent Asset Allocator", advocates overbalancing. It looks like Jason Hsu agrees.

jjface
Posts: 2571
Joined: Thu Mar 19, 2015 6:18 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by jjface » Mon Sep 12, 2016 9:29 am

Interesting. I'll have to read Bernstein's take on it. I've been dabbling with the idea for a while but never had the chance to practice it yet.

User avatar
digarei
Posts: 792
Joined: Sat Jul 05, 2014 1:41 am
Location: Sacramento
Contact:

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by digarei » Mon Sep 12, 2016 9:35 am

"Very small and infrequent policy changes opposite large market moves, more often than not improves overall portfolio performance." - William Bernstein

"If the expected return of an asset class goes up/down, so does [your] allocation to it. If its risk goes up/down, [your] allocation [should be adjusted to] the opposite."

"[The strategy involves] slow and relatively slight changes in allocations based on valuation and expected return."

Rationale provided here:

Mammas, Don’t Let Your Babies Grow Up To Be Timers
Bill Bernstein - Efficient Frontier - William J. Bernstein 2003
http://www.efficientfrontier.com/ef/703/timer.htm

Also, TFB. See the following.

Embrace the Bear Market with Overbalancing
TFB - The Finance Buff - Harry Sit July 15, 2008
http://thefinancebuff.com/embracing-bear-market.html
Connect with Bogleheads in Northern California! Click the link under my user info/avatar.

User avatar
TheGreyingDuke
Posts: 1509
Joined: Fri Sep 02, 2011 10:34 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by TheGreyingDuke » Mon Sep 12, 2016 10:12 am

I like this in theory but in my situation:

1) In draw down phase, no reinvesting of dividends or capital gains

2) 80% of investments are in taxable accounts with significant LT gains

3) Desired allocation is 45%/55% stocks/bonds, currently at 54%/46%

4) I am two years from 70 and collecting SS, filling the 15% bracket with IRA conversions

5) I have some LT capital loss carryover but reluctant to use it while current CH are tax free

Do I bite the tax bullet and re-balance to desired ratio or just let it ride?
Am I better off filling the 15% space with 0% cap gains or with the IRA conversions (ran scenarios and it is not clear to me)?
"Every time I see an adult on a bicycle, I no longer despair for the future of the human race." H.G. Wells

livesoft
Posts: 62925
Joined: Thu Mar 01, 2007 8:00 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by livesoft » Mon Sep 12, 2016 10:22 am

I think one has to market time one's overbalancing moves.

And think about it: Suppose one overbalances by 10% into equities and equities go up 10%. That's an extra 1% performance.

And don't forget that one will not market time perfectly, so sometimes things will go up less and sometimes things will go down. So maybe it averages out to an extra 0.5% performance.

And don't forget that this is not a free lunch because the higher temporary allocation to equities is absolutely riskier.
Wiki This signature message sponsored by sscritic: Learn to fish.

randomguy
Posts: 6621
Joined: Wed Sep 17, 2014 9:00 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by randomguy » Mon Sep 12, 2016 10:59 am

digarei wrote:
Embrace the Bear Market with Overbalancing
TFB - The Finance Buff - Harry Sit July 15, 2008
http://thefinancebuff.com/embracing-bear-market.html
Going stock heavy in July 2008 seems like really bad timing.:) Yes ending up at 85/15 in March was good. But upping by 5% from June to March would have been painful. The question of how well this works will come at the other end. How long do you hold 85/15? Sell after 2 years in 2011, and you miss out on a very good 3 years. Hold too long (say 6 years starting in 2003) and you get hit by the next crash. You need both a strategy to get in an out.

I have seen a bunch of timing strategies. Very few of them increase returns significantly. But decreasing volatility seems possible. How much is curve fitting is hard to say.

Theoretical
Posts: 1439
Joined: Tue Aug 19, 2014 10:09 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Theoretical » Mon Sep 12, 2016 11:06 am

randomguy wrote:
digarei wrote:
Embrace the Bear Market with Overbalancing
TFB - The Finance Buff - Harry Sit July 15, 2008
http://thefinancebuff.com/embracing-bear-market.html
Going stock heavy in July 2008 seems like really bad timing.:) Yes ending up at 85/15 in March was good. But upping by 5% from June to March would have been painful. The question of how well this works will come at the other end. How long do you hold 85/15? Sell after 2 years in 2011, and you miss out on a very good 3 years. Hold too long (say 6 years starting in 2003) and you get hit by the next crash. You need both a strategy to get in an out.

I have seen a bunch of timing strategies. Very few of them increase returns significantly. But decreasing volatility seems possible. How much is curve fitting is hard to say.
I think Bernstein would tend to say you should first focus any moves on overbalancing within a cheaper slice of an asset class (or the broader risk side or the conservative side - corporate/treasury spread) - such as moving from domestic to emerging or developed stocks rather than stocks to bonds.

magneto
Posts: 961
Joined: Sun Dec 19, 2010 10:57 am
Location: On Chesil Beach

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by magneto » Mon Sep 12, 2016 11:20 am

Overbalancing is a name given by Wm B, to what many others will know as Dynamic Asset Allocation, a form of Variable Ratio Formula Investment Plan, which has been around for at least eighty years plus.
And viewed (and employed) with renewed interest by venerable institutions after the 1929 debacle.

More often than not VR relies on awareness of valuations, which for some is the road to ruin!
Most investors here feel far more comfortable with a Constant Ratio Formula Investment Plan.

Always bemused as to why one formula should be damned as market timing, but not the other?
Each is following a formula in a disciplined manner, and each has their own pros and cons.
Why should CR be so sacred?

The problem with VR or even CR is momentum, as posters have picked up on.
Some delaying mechanism, such as bands can ameliorate, but only to a degree.
Momentum will always be the enemy of Value.
'There is a tide in the affairs of men ...', Brutus (Market Timer)

RNJ
Posts: 767
Joined: Mon Apr 08, 2013 9:06 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by RNJ » Mon Sep 12, 2016 11:53 am

Key to remember that what looks cheap/expensive now can become far more so and over a long period of time before any realignment or mean reversion. Tread lightly.

Full disclosure: I do this.

User avatar
whodidntante
Posts: 4307
Joined: Thu Jan 21, 2016 11:11 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by whodidntante » Mon Sep 12, 2016 12:25 pm

I have learned to enjoy the slight sickness that I feel when I buy depressed assets. This is the same approach, but the marketing is better.

User avatar
nedsaid
Posts: 10664
Joined: Fri Nov 23, 2012 12:33 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by nedsaid » Mon Sep 12, 2016 2:24 pm

magneto wrote:Overbalancing is a name given by Wm B, to what many others will know as Dynamic Asset Allocation, a form of Variable Ratio Formula Investment Plan, which has been around for at least eighty years plus.
And viewed (and employed) with renewed interest by venerable institutions after the 1929 debacle.

More often than not VR relies on awareness of valuations, which for some is the road to ruin!
Most investors here feel far more comfortable with a Constant Ratio Formula Investment Plan.

Always bemused as to why one formula should be damned as market timing, but not the other?
Each is following a formula in a disciplined manner, and each has their own pros and cons.
Why should CR be so sacred?

The problem with VR or even CR is momentum, as posters have picked up on.
Some delaying mechanism, such as bands can ameliorate, but only to a degree.
Momentum will always be the enemy of Value.
Brilliant point! We underestimate the power of momentum. This is why I would be very nervous about shorting overvalued stocks. Should work great in theory but momentum can last longer than expected. This is also why Value investing takes patience as Value is uncorrelated to momentum.

Still, I like the idea of selling expensive assets to buy cheaper assets. You will be frustrated if you try to time this exactly. Cheap can get cheaper and expensive can get more expensive.
A fool and his money are good for business.

Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Thu Oct 20, 2016 5:06 pm

It looks like Jonathan Clements also believes in overbalancing. This is from Jonathan Clements Money Guide 2016, p. 206. He states he has about 70% of his portfolio in stocks. Then he states that he would likely increase that to 75% if the market declined 25% or more. He mentions that he owns a REIT fund, though less than his target weighting, as he considers REITs overvalued. He is modestly overweighted in foreign stocks and modestly overweighted in US stocks, relative to his written portfolio targets. The reason is that he considers foreign stocks as better value than US stocks.

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by larryswedroe » Thu Oct 20, 2016 5:18 pm

I also wrote about this in my first version of the Only Guide to a Winning Investment Strategy You'll Ever Need, the 1998 edition
Larry

User avatar
dratkinson
Posts: 4381
Joined: Thu Jul 26, 2007 6:23 pm
Location: Centennial CO

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by dratkinson » Thu Oct 20, 2016 5:46 pm

larryswedroe wrote:I also wrote about this in my first version of the Only Guide to a Winning Investment Strategy You'll Ever Need, the 1998 edition
Larry
Sir, could you provide a clear distinction between "overbalancing" vs. "selling winners hurts momentum", as I've heard you mention both, but the concepts seem to be mutually exclusive.

When is it (in)appropriate to follow one concept or the other?

Should we give more weigh to following one concept than the other?
d.r.a, not dr.a. | I'm a novice investor, you are forewarned.

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by larryswedroe » Thu Oct 20, 2016 6:58 pm

dratkinson
First, rebalancing is done to control risk, not increase returns as some think.

Second, because of both costs (transactions and taxes) as well as momentum, one should set relatively wide rebalancing bands, but not so wide that you end up taking more risk than you have the ability, willingness or need to take.

Third, I didn't recommend the "overrebalancing" though did discuss it as way to possibly increase returns because you are buying what is now much cheaper and selling what is now more expensive. And least WITHIN equity asset classes would consider this, not between stocks and bonds. But you have to be aware that you are taking on more risks when doing it.

Fourth, you have to remember that MOM is a short term phenomenon with long term reversals. in fact AQR in their TS MOM fund uses three signals, a short term and long term and reversal signal. It weights positions based on agreement or disagreement among them. For example, when a signal based on the last 12 months is positive but not so in the last 3 months a TS MOM fund would not weight it so heavily.
Hope that helps
Larry

User avatar
Kevin M
Posts: 10294
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Kevin M » Thu Oct 20, 2016 8:36 pm

digarei wrote: Also, TFB. See the following.

Embrace the Bear Market with Overbalancing
TFB - The Finance Buff - Harry Sit July 15, 2008
http://thefinancebuff.com/embracing-bear-market.html
Harry wrote a number of subsequent blog posts that shared his ongoing results with overbalancing into and out of the financial crisis. To find them, use a search like this:

overbalancing site:https://thefinancebuff.com

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

azanon
Posts: 1992
Joined: Mon Nov 07, 2011 10:34 am
Location: Little Rock, AR

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by azanon » Thu Oct 20, 2016 10:09 pm

.. then you have the momentum and trend guys who will tell you that if you sell the thing that's going up, and buy the thing that's going down, more often than not, you'll start losing. Or if you buy the asset that is steadily dropping in price, it's like trying to catch a falling knife. Momentum is a well documented factor shown to work.

I'm actually not saying which is right or wrong, just pointing out that different experts appear to be saying exactly the opposite things. What are the non-experts to do? Opt-out of all of it, and just buy-and-hold with occasional rebalancing according to a policy statement? :beer

User avatar
digarei
Posts: 792
Joined: Sat Jul 05, 2014 1:41 am
Location: Sacramento
Contact:

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by digarei » Fri Oct 21, 2016 1:38 am

Kevin M wrote:
digarei wrote: Also, TFB. See the following.

Embrace the Bear Market with Overbalancing
TFB - The Finance Buff - Harry Sit July 15, 2008
http://thefinancebuff.com/embracing-bear-market.html
Harry wrote a number of subsequent blog posts that shared his ongoing results with overbalancing into and out of the financial crisis. To find them, use a search like this:

overbalancing site:https://thefinancebuff.com

Kevin

Kevin,

Thanks for alerting me to the follow-on articles. I read all of those flagged by the search, from 2008 through 2011. I was somewhat surprised that he didn't quantify its success in one of the blog posts. I did note his assertion that the outcome of his overbalancing plan was 'good' and that it worked well for him.

"In retrospect, overbalancing worked well for me [...]"

Take Money Off the Table After a Good Run

https://thefinancebuff.com/take-money-o ... d-run.html


digarei
Connect with Bogleheads in Northern California! Click the link under my user info/avatar.

Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Fri Oct 21, 2016 3:51 pm

William Bernstein, in "Rational Expectations" from 2014, mentions overbalancing.

"The data of Dimson, Marsh, and Staunton are highly suggestive that shifting allocations among equity asset classes according to valuation can be beneficial."

'The prime directive for strategic asset allocation is small, infrequent changes in allocation opposite large changes in valuation." Strategic asset allocation is a synonym for overbalancing.

"A practical and conservative way to adjust allocations by price changes might be by, say, a 10-to-1 ratio on the upside and perhaps (because of the lognormal math) a 6-to-1 ratio on the downside."


The following is an aside, but I find it noteworthy. A question about rebalancing in general is whether to use a calendar method versus a threshold method.

Larry Swedroe is an advocate of the threshold method (5/25).

Portfolio Solutions uses a threshold method (https://portfoliosolutions.com/latest-l ... -rebalance), so I assume Rick Ferri advocates that method also.

William Bernstein uses a threshold method (http://blogs.wsj.com/totalreturn/2015/0 ... this-year/).


Once again from "Rational Expectations",

"Because threshold rebalancing tends to “catch” market peaks and valleys more effectively than simple calendar rebalancing, I believe it is likely to be superior to calendar rebalancing. But since rebalancing is a very path-dependent process, I can’t be sure."

"Let’s start with the U.S. large cap component of a portfolio. A reasonable rebalancing threshold might be 20% of the allocation. Twenty percent of a 30% allocation is 6%, so you should rebalance back to target when this asset class hits 36%. Emerging markets are much more volatile, and so a wider threshold is called for here, perhaps 50%. If its policy allocation is 5%, then on the upside you’ll rebalance at 7.5%. On the downside, things are even trickier. Asset returns behave approximately lognormally. A 20% increase exactly offsets a 16.67% decline. A 50% increase offsets a 33.33% decline. Therefore, in the examples above, your targets on the downside should be 25% for U.S. large cap (30% x 0.8333; that is, 30% divided by 1.2), and 3.33% for emerging markets (5% x 0.6667; that is, 5% divided by 1.5). For each individual asset class, these thresholds are works in progress. If you wind up rebalancing an asset class more than once per year on average, you should widen your thresholds. If you’re hardly rebalancing at all, narrow the thresholds. Your target should be to rebalance once every 2 to 4 years for each asset class."

User avatar
digarei
Posts: 792
Joined: Sat Jul 05, 2014 1:41 am
Location: Sacramento
Contact:

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by digarei » Fri Oct 21, 2016 5:15 pm

...
larryswedroe wrote: [...] I didn't recommend the "overrebalancing" though did discuss it as way to possibly increase returns because you are buying what is now much cheaper and selling what is now more expensive. And least WITHIN equity asset classes would consider this, not between stocks and bonds.
Theoretical wrote: Bernstein would tend to say you should first focus any moves on overbalancing within a cheaper slice of an asset class [...]
Park wrote: William Bernstein, in "Rational Expectations" from 2014, mentions overbalancing.

"The data of Dimson, Marsh, and Staunton are highly suggestive that shifting allocations among equity asset classes according to valuation can be beneficial."
  • emphasis applied
William Bernstein and Larry Swedroe are defining Strategic Asset Allocation (aka "over balancing") as valuation-driven changes to allocations WITHIN equities (stocks), and are NOT speaking of changes to allocations between equities and fixed income classes (stocks / bonds). Is this correct?

Reading Bernsteins 2003 article it wasn't clear to me that this was a constraint.

http://www.efficientfrontier.com/ef/703/timer.htm

Its generalized discussion and recount of a conversation he had with Rob Arnott led me to believe, perhaps erroneously, that tactical asset allocation could be applied not merely to sub-asset classes but between asset classes.

"[...] the difference between the expected stock return and that of 10-year treasuries."

"If the expected return of an asset class goes up/down, so does his allocation to it. If its risk goes up/down, his allocation does the opposite."

I wonder whether Harry Sit, back in 2008, interpreted the article the same way I did on first reading?
TFB wrote:Because stocks are cheaper now than what they were a year ago, I’m increasing my stocks allocation from 60% to 65%.
Connect with Bogleheads in Northern California! Click the link under my user info/avatar.

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by larryswedroe » Fri Oct 21, 2016 6:37 pm

Note I'm not advocating that overrebalancing, just stating it is something one could consider, within equities only.
Larry

User avatar
Johnnie
Posts: 512
Joined: Sat May 28, 2016 3:18 pm
Location: Michigan

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Johnnie » Sat Oct 22, 2016 10:13 am

Hmm - it sure feels like timing.

And how would that have worked during something like this?

From "The Worst Bear Market That Nobody Ever Talks About" by Michael Bartnick, 5/4/16, The Irrelevant Investor:
The longest bear market didn’t begin in 1929 or 2007, but rather on January 11, 1973 . The 437 days from peak-to-trough gave birth to many well-known value investors and also left in its wake a generation of brokers that would never return to Wall Street. Roger Lowenstein described how this period is forgotten today.

...Aside from the fact that this was the longest bear market ever, here are five additional reasons why these 21 months were particularly brutal.

"1) It occurred within the context of a massive secular bear market, one that made zero progress over a 16-year period..." (1966-82)...

http://theirrelevantinvestor.com/2016/0 ... lks-about/


Yikes! :shock:
"I know nothing."

Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Sat Oct 22, 2016 2:45 pm

Johnnie wrote:Hmm - it sure feels like timing.
IMO, overbalancing overlaps with tilting to value. Is value investing market timing?

larryswedroe
Posts: 15711
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by larryswedroe » Sat Oct 22, 2016 3:33 pm

now doubt that overbalancing is market timing, which is why I note that if you are going to "sin" sin only a little, don't put too many eggs in any one basket, including the US one
Larry

Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Sat Oct 22, 2016 4:34 pm

From "Rational Expectations" by William Bernstein,

"Calendar rebalancing is still an effective, and quite simple, way to rebalance. If you use this method, do not rebalance more than once a year. Markets tend to exhibit momentum at periods of one year or less, and mean reversion takes place over longer periods. Rebalancing once every two to three years is plenty."

By rebalancing once every 2-3 years to maximize the chance of mean reversion, it reminds me somewhat of trend following. Overbalancing reminds me somewhat of value investing.

https://www.institutionalinvestor.com/a ... ingle=true

Cliff Asness:

"We are going to argue that market timing isn't really a sin except, as for so many things, if done to excess. But the results and logic behind two simple strategies that govern so much of the investing world - basic value, or contrarian, investing and basic momentum, or trend-following, investing - imply that when it comes to market timing, one should indeed sin a little and do so as a matter of course, not just at extremes."

Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Sat Oct 22, 2016 5:17 pm

Rebalancing between stock asset classes is a bet on reversion to the mean. Overbalancing between stock asset classes is a greater bet on reversion to the mean. The hope is that a reversion to the mean will result in a rebalancing bonus. As developed world stock markets have had similar long term returns, it's not been a bad bet to make. What about the future? As developed world stock markets have similar risks, they may very well have similar future returns.

What about stock asset classes with different returns (emerging markets, small cap value etc.)?

http://www.efficientfrontier.com/ef/197/rebal197.htm

The above link, from William Bernstein, would suggest that the difference in returns would have to be greater than 5% for rebalancing not to be beneficial.

User avatar
Johnnie
Posts: 512
Joined: Sat May 28, 2016 3:18 pm
Location: Michigan

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Johnnie » Sat Oct 22, 2016 10:35 pm

Park wrote:
Johnnie wrote:Hmm - it sure feels like timing.
IMO, overbalancing overlaps with tilting to value. Is value investing market timing?
That's correct and a good point. Others said something similar with reversion to the mean.

But here's the difference: With standard fixed-interval rebalancing to your predetermined A/A you never have to make any decisions or judgment calls. That's the idea of Boglehead investing, in fact, right? Set it and forget it, barring civilizational-level disturbances.

I suppose you can write a series of if-then rules, but even that feels more like something a hedge fund would do (sorry, that could be an insult but isn't intended that way. ;-) )
"I know nothing."

Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Sun Oct 23, 2016 10:25 am

https://blogs.cfainstitute.org/investor ... rebalance/

The above link is from Jason Hsu:

"Statistically, there is documented intermediate-horizon mean reversion in equity returns and long-term mean reversion in asset class returns...

Two features work in your favor when applying contrarian rebalancing within asset classes: (1) shorter mean-reversion horizon and (2) a larger cross-section. Mean reversion is a very noisy signal, thus you really need a lot of securities to make the effect work reliably. When you aggregate the effect across hundreds of securities within an asset class, the law of large numbers kicks in to wash out the noise and accentuate the mean-reversion effect.

When applying contrarian rebalancing across asset classes, if you don’t have many distinct asset classes, the benefit would be more lumpy. Additionally, since the asset class mean-reversion horizon is a bit longer, you might have to wait a bit for the effect to really kick in and work for you...The classic argument of rebalancing to, say, a 60/40 portfolio, is more troublesome. You only have two asset classes, so you don’t have the law of large numbers on your side. The asset class mean reversion also takes place over a much longer horizon. We are talking about a minimum of five years. So at that level, if you try to measure rebalancing benefit, I’m not surprised that most wouldn’t find satisfying evidence."



http://www.morningstar.com/Cover/videoC ... ?id=716560

The above link is from Frank Kinniry of Vanguard:

"Ptak: Why do you think that making tilts at the asset class level is perhaps less hospitable to an active manager than maybe bottom-up security selection would be?

Kinniry: I think the key is back to probability theory. If you buy 400 or 500 or 800 stocks, and you're plus or minus 10 basis points from its market cap-weight, hypothetically, if you get 55% or 60% of them right, you're going to outperform relative to your costs.

But if you tilt a portfolio, let's say, 5% more aggressive throughout the last 15 years, or 5% more conservative, the dispersion of stocks and bonds and the correlation of them are so wide. So, you end up with even a small bet or a small overweight/underweight blowing out the distribution. Whereas with securities, they tend to correlate pretty closely. So you're getting the overall dispersion down, and we think the best chance for active success.

Ptak: Last question: Has any of Vanguard's research shown that a tactical asset allocation strategy can perhaps nicely complement a more traditional portfolio allocation to stocks and bonds?

Kinniry: Just like we really believe in active management, if we felt there was a strong probability of us being successful, we probably would implement it. So it's not without a lot of research and a lot of effort in trying, and we'll continue to try. But as of now, we really haven't found that tactical asset allocation makes a lot of sense for our assets in the portfolios."



http://www.morningstar.com/cover/videoc ... ?id=353908

The above link is a comment from Frank Kinniry of Vanguard on market timing. Based on what he says below, he might not be in favor of overbalancing between equity asset classes, although Cliff Asness might be.

"Kinniry: We've looked at it, long and hard. And actually, we have a real reason to look at this and spend a lot of time looking at it, and really the only thing we've found that has some merit is at the stock/bond mix, meaning we don't see a lot between U.S. and international or growth and value or size, but we do see some ability of tactical allocation in the extremes and really the extremes are limited periods in time where this occurs. So, 1998-1999, when you saw a valuations at unprecedented levels, one could have said the equity market was set up for potentially lower returns, and in 2009, believe it or not, at the bottom, the equity market looked about as attractive as it had in 20 to 25-plus years.

And so there is a lot of noise in the middle, but there are some small windows in time where the stock/bond mix, you may want to shade, and we would say only small shades if you were 60/40 stock/bonds, maybe it would go up 5% or 10% in your equities or down 5% or 10%, but really never making wholesale moves.

Benz: And arguably a rebalancing strategy would kind of get you there.

Kinniry: Absolutely, a rebalancing strategy is probably the best way to take advantage of some of these opportunities without making wholesale changes; that's exactly right."

(The following is a tangent, but an interesting one IMO:

Larry Swedroe:

"I agree that market timing at EXTREMES is a good idea, but only if you are prepared to
a) be wrong for long time (I sold all growth in 98 and looked bad for 2 years, then looked great)
b) don't confuse strategy and outcome
c) stay disciplined (the hard part is when do you get back).

so for most, especially if have well diversified by asset classes/factors, simply rebalancing is likely to prove better.

and I've only made a few trades over the past 20 years including getting out of growth in 98 and out of REITS a few years ago. Valuations do matter, and they matter a lot."

It looks like Frank Kinniry thinks you can't market time within asset classes. But Larry Swedroe thinks you can, at extremes of valuation. I think Larry is right.)




However, what is written below is from the first interview in this post with Frank Kinniry:

"Ptak: Vanguard at one time had its own dynamic asset allocation strategy; it was the Asset Allocation fund, which was merged into another of your offerings back in 2011. Can you take us through the thought process that informed the decision to merge that strategy away, and maybe any implications that it might have on an investor who was considering a strategy like it for their own use?

Kinniry: I would say we mostly were happy with the performance of the fund. The fund was around for over 20 years and actually had large assets under management. And I think the good thing at Vanguard, what we always do, is we are diligent about making sure that our funds still meet the requirements for investor success. At Vanguard, we actually have a core purpose, and that's to give investors the best chance of success. And we look at some of our funds, even though the performance may be good, and we ask, "Are investors using this to their own benefit?" And after looking at all that, we came to the conclusion that while it was a good offer, it had done well, and had large assets, we decided it would be best just to go into a strategic asset allocation. So we wound that fund into the Balanced Index Fund"

Vanguard had a market timing fund (Asset Allocation fund) from 1988-2011. The fact that it had a market timing fund for 23 years says something. So there's a bit of a mixed message here. And this fund could switch between US stocks, bonds and money market instruments: it wasn't limited to switching between equity asset classes.





What do I conclude about overbalancing? Do it only within your equity asset classes. Do it modestly and have modest expectations. And the links in this post would indicate that the more equity asset classes you overbalance between, the better. This would also give you greater control over your diversification. On the other hand, the more equity asset classes you have, the more costs there are associated with overbalancing (time, transaction costs and taxes).



Edited to include the following:

As mentioned previously, rebalancing is a bet on reversion to mean. So rebalancing is contrary to momentum. This applies to rebalancing between equity asset classes. That case is not as strong for rebalancing between stocks and bonds, where over the long run, stocks should outperform stocks.

Campbell Harvey has some issues, when it comes to rebalancing. See the link below:

http://www.marketwatch.com/story/the-hi ... 2014-12-09

"Whenever the market is in a longer-term up or down trend, rebalancing actually increases risk, particularly downside risk. That’s because, when you rebalance, you take money away from the better-performing asset class and reinvest it in the poorer-performing one. If those two asset classes’ relative strength persists after the rebalancing, as they often do, you’ll end up worse off than if you had not rebalanced."

So when you rebalance, you're making a bet against momentum, and that instead there will be a reversion to the mean. Sometimes that bet will be right and it will decrease risk, but not always. By rebalancing, you're taking on the risk that you'll make a protracted bear market worse. And decreasing such risk is what many want to do, when they rebalance stock/bond portfolios.

Campbell Harvey believes the decision about that bet "depends on your tolerance for risk and the time horizon of your investments. If you are willing to undertake more risk and have a very long-term horizon of at least 10 to 20 years, then rebalancing is an appropriate strategy. That’s because, over long periods of time, a periodically rebalanced portfolio is likely to earn enough to compensate you for its greater risk.

However, Harvey said, rebalancing becomes less appropriate to the extent you are a conservative investor and have an investment horizon shorter than 10 years...If you nevertheless insist on doing so, Harvey said you should not mechanically rebalance at the end of every quarter or year, but instead time your rebalancings for periods when the markets’ trends appear to be reversing.

If you rebalance every 2-3 years, you'll be going against momentum less, and making more of a bet for mean reversion. The following is relevant.

AQR has a paper on rebalancing.

https://www.aqr.com/library/aqr-publica ... allocation

The paper gives autocorrelations for asset classes between 1972-2014.

Positive autocorrelation favors momentum. Negative autocorrelation favors reversion to mean.

US stocks: 12 months 0.05, 3 years -.25, 5 years -0.13
NonUS stocks: 0.11, -0.37, -.45
US Bonds: 0.21, -0.18, 0.01
NonUS Bonds: 0.12, --0.51, -0.65

So based on this, waiting for 3 years to rebalance makes sense.

http://www.efficientfrontier.com/ef/100/rebal100.htm

William Bernstein looked a rebalancing a portfolio of 40% S&P500, 15% US small stocks, 15% foreign stocks and 30% 5 year government bonds.

Rebalancing monthly resulted in a return of 12.030%, quarterly 12.090%, every year 12.091%, every 2 years 12.166%, every 4 years 12.267%. The differences are small, although costs are ignored, and less frequent rebalancing might decrease costs.

https://www.pwlcapital.com/pwl/media/pw ... f?ext=.pdf

The link above is to a paper which looked at rebalancing several different stock/bond portfolios that an American, Japanese, Canadian or British investor might have. Results are given for 1 year rebalancing, 2 years, 3 years and 5 years; costs are ignored. The differences between rebalancing periods are very small. There might/might not be a very slight benefit to waiting to 3 years. But the longer you wait, the more concentrated and less diverse your portfolio becomes relative to the target asset allocation.

About overbalancing, it once again makes your portfolio more concentrated and less diverse relative to the target asset allocation. However, overbalancing will tend to tilt to value. One can make an argument that such a tilt will decrease the risk of a portfolio

Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Tue Oct 23, 2018 8:24 am

https://www.morningstar.com/videos/8871 ... nment.html

My impression is that Joe Davis, Vanguard's chief economist might like overbalancing too:

"If one is comfortable taking some, I'd call it, active risk relative to their policy or portfolio--let's just say with 60% stocks, 40% bonds, Christine--if they are willing to tolerate periods of underperformance, that's at least a framework I would be looking through to say, if I'm going to tilt my portfolio. Personally, I'm one of those investors. What I tend to do is, say, what are valuations in the market on a smooth, long-term basis. That would tell me that I should be rebalancing my portfolio and even potentially putting new money into work, into overseas markets, into value, parts of the market that have underperformed.

I know personally I will never get the timing right. It's a long-term, discretionary or automatically rebalancing to those parts of the market that have just underperformed. That's a nice disciplined approach. Any one year, it may not seem like it's paying off. But I do know from the research that over--and you've done some as well, Christine--over 10- or 15-, 20-year period you can add, I don't know, roughly 50 basis points in, I'd call it, rebalancing alpha. You could call it, slightly tilting one's portfolio. But one has to stick with it and be systematic in it."

Morse Code
Posts: 112
Joined: Wed Mar 14, 2007 12:02 pm
Location: Grand Rapids, MI

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Morse Code » Tue Oct 23, 2018 8:54 am

This glide path method does it systematically, without the need to make decisions. Just rebalance according the formula and you automatically "over-rebalance".

viewtopic.php?f=10&t=261205&hilit
Livin' the dream

Ron Scott
Posts: 1090
Joined: Tue Apr 05, 2016 5:38 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Ron Scott » Tue Oct 23, 2018 9:07 am

It is interesting to observe the linguistics needed to instruct the religious how to sin.

"Overbalancing" is by definition market timing but some are so afraid to advocate it or do it they need to dance around it to ease their minds.

Hsu: "take money away from what's done really well and become expensive and then move money into what's done poorly...So if you do this, you know, a systematic rebalancing, or maybe do a little bit more, a systematic over-rebalancing, you could develop an edge certainly, over people who are more trend-chasing in how they make decisions"
Retirement is a game best played by those prepared for more volatility in the future than has been seen in the past. The solution is not to predict investment losses but to prepare for them.

Bill Bernstein
Posts: 562
Joined: Sat Jun 23, 2007 12:47 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Bill Bernstein » Tue Oct 23, 2018 10:22 am

I confess! "Overbalancing," AKA "Strategic AA," *is* market timing.

But so, then, is simple constant AA rebalancing; it's just a matter of degree.

Both are simply a bet on mean reversion, and neither will work in a truly random walk world.

Whether that's a bet worth making is up to the individual.

Bill

staythecourse
Posts: 6204
Joined: Mon Jan 03, 2011 9:40 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by staythecourse » Tue Oct 23, 2018 10:36 am

This is why I love this board. Bring up a topic written by a couple great authors and there we go BOTH show up and give input. Thanks Mr. Swedroe and Dr. Bernstein.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

Random Walker
Posts: 3316
Joined: Fri Feb 23, 2007 8:21 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Random Walker » Tue Oct 23, 2018 10:49 am

Hi Bill,
Thanks for chiming in! I think it’s a good time to mention your comments at the end of you Life Cycle Investing book. I’ve taken to heart Murphy’s Law of Retirement and your comments on sort of customizing one’s own glide path towards the retirement portfolio. A swollen portfolio results from high past returns. These high past returns are due in large part to increasing valuations. Increased valuations imply lower future expected returns. In fact the whole potential future dispersion of returns shifts left. Makes sense to take a significant amount of risk off the table when one is within sight of retirement under these conditions. I like the idea of opportunistically making steps towards the retirement portfolio rather than blindly taking 1-2% per year off according to age. When the SD of equities is about 20%, seems there should be opportunities to do a bit better than a target date fund. Of course this requires one to have a good feel for their realistic financial goals, and to strongly distinguish between needs and wants. I’m a huge fan of viewing the portfolio as a whole from the top down. But I’ve really come to appreciate the concept of a liability matching portfolio and a risk portfolio which I first learned from Dr. Bernstein. Makes sense to take some definitive steps to secure the LMP.

Dave

JW-Retired
Posts: 6998
Joined: Sun Dec 16, 2007 12:25 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by JW-Retired » Tue Oct 23, 2018 10:52 am

Overbalancing sounds like a lot of work! :D

Personally, I don't understand why more folks don't do what I choose follow, which is do only one-way bull market re-balancing? That forces you to sell your existing portfolio stocks only when you exceed your upper stock band near market highs, and never at anything but that. In a bear market it lets your skip all the work of selling bonds & buying stocks to re-balance when they are going down, and then reversing that as they are going back up.

I don't really see how the one-way re-balancing could be worse than two-way band re-balancing or overbalancing, and it's so very much less work and worry. Also good insurance against a long term 1930's type disaster.

I've done this one-way thing for the last 20 years and it seems to have produced good results. Can anyone point to a careful analysis of it versus the standard two-band stuff?
JW
Retired at Last

Bill Bernstein
Posts: 562
Joined: Sat Jun 23, 2007 12:47 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Bill Bernstein » Tue Oct 23, 2018 11:11 am

What?? Give up the chance to buy low?? Perish the thought.

Seriously, though, it's all a continuum; one-way rebalancing is just half-way in between normal constant allocation rebalancing and never rebalancing.

And in retirement, it's maybe only 1/4 the way towards never rebalancing, since by definition in the decumulation phase, one is, on average, to keep a constant AA, doing a lot more selling of equities than buying; the reverse is true during the savings phase.

I doubt you'll find any good data on one-way rebalancing, since because the process is so path-dependent, and the return differences so small, there are no reliable data even on more conventional rebalancing strategies (for example, threshold vs calendar) to begin with.

Bill

balbrec2
Posts: 119
Joined: Mon Nov 13, 2017 3:03 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by balbrec2 » Tue Oct 23, 2018 1:34 pm

Park wrote:
Mon Sep 12, 2016 9:21 am
http://news.morningstar.com/cover/video ... ?id=768821

Hsu: "take money away from what's done really well and become expensive and then move money into what's done poorly...So if you do this, you know, a systematic rebalancing, or maybe do a little bit more, a systematic over-rebalancing, you could develop an edge certainly, over people who are more trend-chasing in how they make decisions"

Kinnel: "you're suggesting you could even go another step further and even add a little more to the worst-performing investments in your portfolio, take a little more away from the strongest performers"

Hsu: "That's absolutely true, and like you say, within a tax-advantaged account, the benefit of rebalancing is phenomenal because it doesn't have the tax consequence associated with it. I've even seen research, where careful and infrequent rebalance is still a lot better even in the face of some tax consequences."

William Bernstein, in "The Intelligent Asset Allocator", advocates overbalancing. It looks like Jason Hsu agrees.
Over rebalancing smacks of Lichello's AIM theory!

asif408
Posts: 1500
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by asif408 » Tue Oct 23, 2018 2:36 pm

JW-Retired wrote:
Tue Oct 23, 2018 10:52 am
Overbalancing sounds like a lot of work! :D
Not really. Just buy a little more of the laggards or sell a little more of the winners. For example, I just overbalanced and bought more emerging markets and precious metals about a month ago. They are down 20-30% since January, and that's after a decade of double digit negative returns for many of these markets while the S&P doubled.
JW-Retired wrote:
Tue Oct 23, 2018 10:52 am
I don't really see how the one-way re-balancing could be worse than two-way band re-balancing or overbalancing, and it's so very much less work and worry. Also good insurance against a long term 1930's type disaster.
I'm not sure why there wouldn't be a benefit both ways, if there is, so if you eliminate one it will probably do worse that both. But if your emotional state is such that you'd sell out instead of buying low, then yes, you should stick with your strategy. I have a high enough level of emotional detachment from my investments that it doesn't bother me to buy more of what's done bad even if it keeps going down for a while after I buy more. Sometimes I wonder if I have a mild case of Asperger's......

Dudley
Posts: 103
Joined: Tue May 09, 2017 5:34 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Dudley » Tue Oct 23, 2018 3:01 pm

Bill Bernstein wrote:
Tue Oct 23, 2018 10:22 am
I confess! "Overbalancing," AKA "Strategic AA," *is* market timing.

But so, then, is simple constant AA rebalancing; it's just a matter of degree.

Both are simply a bet on mean reversion, and neither will work in a truly random walk world.
That answer hits the nail on the head.

User avatar
JoMoney
Posts: 6284
Joined: Tue Jul 23, 2013 5:31 am

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by JoMoney » Tue Oct 23, 2018 3:17 pm

Regarding 'One way rebalancing'
This paper, sort of, has a study on it in relation to the Buffett 90/10 portfolio:
Buffett's Asset Allocation Advice: Take it ... With a Twist
https://papers.ssrn.com/sol3/papers.cfm ... id=2680084

He compares the portfolio of a 90/10 regularly rebalanced to one in which two twists are applied, that are essentially similar to "one way rebalancing"
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

Park
Posts: 566
Joined: Sat Nov 06, 2010 4:56 pm

Re: Jason Hsu and William Bernstein Like Overbalancing

Post by Park » Tue Oct 23, 2018 11:53 pm

JoMoney wrote:
Tue Oct 23, 2018 3:17 pm
Regarding 'One way rebalancing'
This paper, sort of, has a study on it in relation to the Buffett 90/10 portfolio:
Buffett's Asset Allocation Advice: Take it ... With a Twist
https://papers.ssrn.com/sol3/papers.cfm ... id=2680084

He compares the portfolio of a 90/10 regularly rebalanced to one in which two twists are applied, that are essentially similar to "one way rebalancing"
Here is a link to a summary of the above paper:

https://seekingalpha.com/article/362334 ... ice?page=3

The two twists are dynamic asset allocation strategies.

Post Reply