Cut-Throat, I keep my stock allocation relatively low (roughly 35% now, but I will go as high as 50% or so if stocks get creamed). So I'm always quite a long ways away from having to sell stocks to pay the bills. Therefore I'm not sure volatility is as big an issue for me as it would be for someone with a high stock allocation, otherwise I'd jump to the Bernstein model. I just can't decide if a switch is worth the hassle. I keep going back to Sharpe - that any diversion from the Total Market approach is a bet, and I have no strong opinions driving a desire to make any bets, and I'm not sure volatility is a big issue for me. Does this make sense? Can you (or anyone else) think of any other reasons to change horses? I'm suffering from analysis paralysis!! :-/
Safe Withdrawals of up to 6% per year? Part 1
- Thread starter LawGirl
- Start date
C
Cut-Throat
Guest
Bob,
I think sticking to a plan is worth probably more than which plan you go with. Berstein points that out. If you start switching gears, you let emotions get the best of you (one of the 4 pillars).
I think investing in the total stock market is fine. I would couple that with other asset classes such as REIT's, Bonds, and Foreign Stocks, Emerging Markets.
I think Unclemick's approach is the simplest with his Vanguard Retirement Portfolio. I went with individual asset classes, because I don't mind putzing a bit and this fufills some of the male testosterone thing that Unclemick refers to.
But I think developing a plan and Sticking to it is the best thing you can do.
I think sticking to a plan is worth probably more than which plan you go with. Berstein points that out. If you start switching gears, you let emotions get the best of you (one of the 4 pillars).
I think investing in the total stock market is fine. I would couple that with other asset classes such as REIT's, Bonds, and Foreign Stocks, Emerging Markets.
I think Unclemick's approach is the simplest with his Vanguard Retirement Portfolio. I went with individual asset classes, because I don't mind putzing a bit and this fufills some of the male testosterone thing that Unclemick refers to.
But I think developing a plan and Sticking to it is the best thing you can do.
Unclemick, I'm not looking for the best performance. I'm just looking at the pros and cons and trying to make a decision. Let me ask you this? Discipline aside, do you see other disadvantages/advantages to Bernstein's approach?
Hyperborea
Thinks s/he gets paid by the post
And a well constructed one can provide better insurance against risks than a straight S&P500/Wilshire5000/US market only portfolio. It won't track the same as one either so that might be scary for some who want to compare to others but over the longer run it should be comparable unless of course the need for that "insurance" occurs.
Cut-throat and unclemick, I think what it comes down to is that there are very compelling arguments for both approaches. I have tried to tip the scales one way or the other, and it always comes up 50-50. It helps to pick two of the best brains out there. The fact that you each have chosen a different approach confirms my thinking, and explains my paralysis. I'm sticking with the same horse. Thanks for hashing it out with me.
Hyperborea
Thinks s/he gets paid by the post
What confuses me is that those pushing "total market portfolios because otherwise you are making bets on sectors" just about always mean total US market. The problem with that is that you are making country bets - it might pay off and it might not. If you are talking about a total world market portfolio, as Sharpe appears to be (I'm only in the 1st of the the 3 lectures though I've peeked ahead to see the ending), then that has some validity. I can see the reasonable argument for mixing a total world equity portfolio in some %age with some %age of bonds to set the desired risk level. At rough national market caps that would mean that ~40% of the equities would be in the US market and 60% in the rest of the world.
There is some evidence for value and small premiums though there is only some evidence and it isn't definitive. That might suggest that one could take a "bet" by including an extra weighting to those sectors though I wouldn't make it a large bet.
As I mentioned before I think Bernstein (though I'm not sure if it was him nor can I re-find the reference) did a study using international market data and built such a national market cap weight portfolio (the total world portfolio) and rebalanced it yearly to account for changing market caps. IIRC he started this portfolio simulation sometime in the late 1800's and ran it until 2000 or so. The results were that you would have done just about as well with this portfolio as you would have by "magically" choosing the equity market that would have done the best. We can't know which one will be the best over the future so by choosing only one national market or overweighting one national market we are taking a bet.
A simple "total world portfolio" could be something like 40% VTSMX + 60% VGTSX. For those not using Vanguard perhaps something like 40% VTI + 50% EFA + 10% EEM though I may have EEM set too high as this is just off the top of my head. This portfolio would then make up the equity portion with bonds mixed in to bring the desired risk level into line.
For those who imagine that the US ceeding the leading country position has to be anything at all like a "fall of the Roman Empire" scenario just look at the last such transition. Britain had no great collapse. The US passed British GDP sometime around 1870 or so and China is expected to do the same to the US in about 10-20 years. For the very old "early retirees" in their 60's or 70's then this transition may have little to no effect on them. Those of us in our 20's, 30's and 40's should be thinking carefully about better diversification - a total world portfolio at least.
Yes Hyper, I am referring to a total world market. But I'll probably maintain some home bias.
Do you recall where you saw the study showing that a world market cap weight portfolio would have done just about as well as "magically" choosing the equity market that would have done the best. I'd be interested in seeing that.
unclemick
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
One small concern with slice and dice beyond discipline and any costs related to rebalancing is turnover.
Recently - Vanguards small cap value index has done well with over 13%/yr the last five years. 0.27% expense ratio BUT 109% turnover. Back when I last looked it was 28%.
The Boglehead in me - says this is costing something. I've wondered about slicing asset classes so thin that 'internal in and out of asset class' is generating trading costs some may not be aware of.
As for U.S./rest of the world cap weight - at age 61 and no plans to become an ex pat, I'm not as concerned - having been thru the 70's and early 80's - hopefully I know the drill.
That said - having taken a lot of ribbing from Brits/Commonwealth type over thirty years - myopic, proventcial, etc. The rest of the world market agruments deserve attention as previously mentioned - if you plan to live abroad, have a long ER span, or just want to be aware of what's availible 'out there'.
Recently - Vanguards small cap value index has done well with over 13%/yr the last five years. 0.27% expense ratio BUT 109% turnover. Back when I last looked it was 28%.
The Boglehead in me - says this is costing something. I've wondered about slicing asset classes so thin that 'internal in and out of asset class' is generating trading costs some may not be aware of.
As for U.S./rest of the world cap weight - at age 61 and no plans to become an ex pat, I'm not as concerned - having been thru the 70's and early 80's - hopefully I know the drill.
That said - having taken a lot of ribbing from Brits/Commonwealth type over thirty years - myopic, proventcial, etc. The rest of the world market agruments deserve attention as previously mentioned - if you plan to live abroad, have a long ER span, or just want to be aware of what's availible 'out there'.
Bob_Smith, et. al,
I have mentioned this before but let me repeat
myself. I use the coffeehouse approach which
equally weights all equity asset classes. I cheat
a little by using 7 equity classes instead of the
usual 6. The extra class is small cap international
represented by Vanguard's International Explorer.
This gives an over all exposure of about 28%
of my equity to international.
The reason I am doing this is to avoid the reverse
DCA affect of drawing down a balanced fund like
Target Retirement 2025. At age 70 with RMD
in force, I feel that I can manage the distribution
more efficiently. I remains to be seen if I can
rebalance during severe stock market down years.
That, they say, is when the rubber hits the road.
For those in the accumulation phase, I think unclemick's
balanced index with a splash of REIT is just fine. I
would add a dash of international as well, however.
Cheers,
Charlie
I have mentioned this before but let me repeat
myself. I use the coffeehouse approach which
equally weights all equity asset classes. I cheat
a little by using 7 equity classes instead of the
usual 6. The extra class is small cap international
represented by Vanguard's International Explorer.
This gives an over all exposure of about 28%
of my equity to international.
The reason I am doing this is to avoid the reverse
DCA affect of drawing down a balanced fund like
Target Retirement 2025. At age 70 with RMD
in force, I feel that I can manage the distribution
more efficiently. I remains to be seen if I can
rebalance during severe stock market down years.
That, they say, is when the rubber hits the road.
For those in the accumulation phase, I think unclemick's
balanced index with a splash of REIT is just fine. I
would add a dash of international as well, however.
Cheers,
Charlie
Hyperborea
Thinks s/he gets paid by the post
I wish I did but I've tried to find it before and haven't been able to - perhaps it was one of the other Bernstein's books (Peter Bernstein)?
I just spent a bit of time searching though and I came up with this interesting paper that I've only had a chance to skim but it appears to support this.
http://www.gsm.uci.edu/~jorion/papers/century.pdf
From page 18 we have:
There was a reference to this paper and some more discussion of the US equity premium in a paper by Jeremy Siegel on "The Shrinking US Equity Premium".
http://www.jeremysiegel.com/view_article.asp?p=171&h=1#_ftnref8
Charlie - or anyone - I'm in the withdrawal phase and my stock allocation is relatively low. I plan to rebalance annually, and because my stock allocation is low (and will remain so) I doubt that I will need to sell stock funds to pay the bills. So I shouldn't encounter reverse dollar cost averaging or the need to sell stocks when they're down. My point is this: doesn't a slice and dice portfolio make the most sense for those with relatively high stock allocations? In other words, doesn't a lower stock allocation reduce the impact of volatility all by itself, and eliminate that as a reason to slice? Is my thinking right on this, or am I missing a piece of the puzzle?
C
Cut-Throat
Guest
Bob,
As I understand it the bond portion of the portfoilo is to reduce volatility. The slice portion of the stock is mainly to increase return over the long haul. Even though some asset classes do tend to reduce volaltility.
Someone correct me, if I'm wrong.
Cut-Throat, that's the way I understood it too. Charlie mentions the problem of reverse DCA, which is important. It isn't an issue for those in the accumulation phase, but it could cut a few years off the life of a portfolio for someone in the withdrawal phase. Slicing helps with the problem. It enables one to withdraw from the slices that have appreciated to restore the balance. But if one has all their stock holdings in a total market fund, you're forced to sell a piece of all the slices because they are all blended together in one fund - you can't pick and choose.
I'm thinking that shouldn't be an issue for those who maintain a relatively low stock allocation because in practice, they shouldn't ever be in a position where they need to sell when prices are down. In other words, the "insurance" Hyper mentions shouldn't be an issue for someone with a stock allocation of 35-40%. Then it just comes down to which is likely to perform better. When you factor in the simplicity and the reduced transaction costs on one hand vs. the chance you may do better with the fine tuning one can do with slicing - it seems to be a coin toss.
I'm thinking that shouldn't be an issue for those who maintain a relatively low stock allocation because in practice, they shouldn't ever be in a position where they need to sell when prices are down. In other words, the "insurance" Hyper mentions shouldn't be an issue for someone with a stock allocation of 35-40%. Then it just comes down to which is likely to perform better. When you factor in the simplicity and the reduced transaction costs on one hand vs. the chance you may do better with the fine tuning one can do with slicing - it seems to be a coin toss.
Bernstein and Swedroe both argue that slicing your
stock allocation will help reduce the volatility of
your port. The "total market" is dominated by
large cap growth stocks. When they are out
of favor, your port would suffer if that was all
you had.
Bob_Smith, IMHO you would still have a lower overall
volatility by slicing even if you had a relatively low
stock allocation. Take today for example. My
coffeehouse was down .59% but every one of
Vanguard's 60/40 funds that use TSM was down
even more. My coffeehouse exposure to international
small cap bucked the trend.
Cheers,
Charlie
stock allocation will help reduce the volatility of
your port. The "total market" is dominated by
large cap growth stocks. When they are out
of favor, your port would suffer if that was all
you had.
Bob_Smith, IMHO you would still have a lower overall
volatility by slicing even if you had a relatively low
stock allocation. Take today for example. My
coffeehouse was down .59% but every one of
Vanguard's 60/40 funds that use TSM was down
even more. My coffeehouse exposure to international
small cap bucked the trend.
Cheers,
Charlie
wabmester
Thinks s/he gets paid by the post
- Joined
- Dec 6, 2003
- Messages
- 4,459
There's a reason for that. Large caps are large caps because that's the way "market experts" have allocated their capital. To overweight small caps is betting that you are smarter than the market -- it's really as simple as that.
I don't recall Bernstein ever discussing the Efficient Market Hypothesis, but it's pretty well accepted, and the whole slice and dice approach he espouses goes against EMH.
Not that there's anything wrong with that. I don't believe the market is all that smart either
Wab, the "lumpers" and "splitters" have been
at odds on this since day 1. According to the
experts, most of the heavy lifting is done when
you decide on your stock/bond/cash allocation.
Over the long term, both approaches will have
close to the same compound annual return.
Over shorter durations, the "splitter" approach
avoids having most of your eggs in the large
cap growth sector, which can be out of favor
for several years at a time. Of course you miss
part of the big returns when large cap growth
is in the driver seat. On balance, equally
weighting the 4 corners plus REIT and International
just seems safer to me .......... and at age 70, my
horizon is a lot shorter than most of the great
unwashed who may read this.
Cheers,
Charlie
at odds on this since day 1. According to the
experts, most of the heavy lifting is done when
you decide on your stock/bond/cash allocation.
Over the long term, both approaches will have
close to the same compound annual return.
Over shorter durations, the "splitter" approach
avoids having most of your eggs in the large
cap growth sector, which can be out of favor
for several years at a time. Of course you miss
part of the big returns when large cap growth
is in the driver seat.
On balance, equallyweighting the 4 corners plus REIT and International
just seems safer to me .......... and at age 70, my
horizon is a lot shorter than most of the great
unwashed who may read this.
Cheers,
Charlie
wabmester
Thinks s/he gets paid by the post
- Joined
- Dec 6, 2003
- Messages
- 4,459
Charlie, I'm not saying one camp is wrong and the other is right. I'm just trying to point out some fundamental philosophical underpinnings of the two approaches that most people miss.
Assuming "lumper" means somebody who buys the world market using cap-weighted indices that match the way the world allocates capital, then a lumper is simply following the market and betting that the net effect of all investors is to create an efficient market with the proper capital allocation. This is a very simple approach, and it doesn't require any rebalancing if you maintain a representitive sample of the market in your portfolio. It also leverages the efficiencies inherent in the world markets (including US, of course).
"Splitters" are betting that historical correlations between "asset classes" are what matter. They disregard how the experts of the world have allocated their capital, and they pick a fairly arbitrary allocation based on voodoo and occassionally MVO tools.
Both groups are probably working under false assumptions, but I think the splitters have a tougher time justifying their allocations than the lumpers.
Consider this analogy (reductio ad absurdum):
A stock market only has two companies: GEE and LEM. GEE is a $100,000,000 cap company (large cap), and LEM is a $100 cap lemonaide stand (small cap).
Two investors each have $1M to allocate. The lumper will allocate $1 to the lemonaide stand, and $999K+ to GEE.
The splitter will invest $500K in each.
Which investor is more rational?
Assuming "lumper" means somebody who buys the world market using cap-weighted indices that match the way the world allocates capital, then a lumper is simply following the market and betting that the net effect of all investors is to create an efficient market with the proper capital allocation. This is a very simple approach, and it doesn't require any rebalancing if you maintain a representitive sample of the market in your portfolio. It also leverages the efficiencies inherent in the world markets (including US, of course).
"Splitters" are betting that historical correlations between "asset classes" are what matter. They disregard how the experts of the world have allocated their capital, and they pick a fairly arbitrary allocation based on voodoo and occassionally MVO tools.
Both groups are probably working under false assumptions, but I think the splitters have a tougher time justifying their allocations than the lumpers.
Consider this analogy (reductio ad absurdum):
A stock market only has two companies: GEE and LEM. GEE is a $100,000,000 cap company (large cap), and LEM is a $100 cap lemonaide stand (small cap).
Two investors each have $1M to allocate. The lumper will allocate $1 to the lemonaide stand, and $999K+ to GEE.
The splitter will invest $500K in each.
Which investor is more rational?
unclemick
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Heh,heh,heh
Also balanced index -to the extent they leave well enough alone and let the computers rebalance automatically in rising and falling markets.
Also balanced index -to the extent they leave well enough alone and let the computers rebalance automatically in rising and falling markets.
Hyperborea
Thinks s/he gets paid by the post
A question on the "world portfolio" though is how far do you follow the world division of investment dollars? It's not all in equities but is partially tied up in bonds, real estate, etc. Do you divide your portfolio totally as the market has?
wabmester
Thinks s/he gets paid by the post
- Joined
- Dec 6, 2003
- Messages
- 4,459
I don't think EMH can help you much with allocation across asset classes (and I don't consider the arbitrary lines drawn between small/mid/large/value/growth stocks to define distinct asset classes).
I'm not even sure data is available that would tell you how the world's investors allocate between stocks, bonds, real estate, metals, etc. Stocks are the only class that is held pretty much exclusively for investment. Metals and real estate are skewed by their utility value. And bonds are held for all sorts of weird reasons -- everything from governments trying to control the strength of their currency to giant mortgage companies hedging their loans.
I use bonds to dampen portfolio volatility and provide guaranteed income, so I tune my allocation to suit my needs (and volatility fears).
I use real assets to ensure that I'm always holding something that will continue to have fungibility in the face of just about any economic storm. So, I adjust my allocation based on my worst-case predictions for other assets.
I guess I take some solace in the lack of historical covariance among these assets too, but I don't really use that to help with allocation decisions.
I'm not even sure data is available that would tell you how the world's investors allocate between stocks, bonds, real estate, metals, etc. Stocks are the only class that is held pretty much exclusively for investment. Metals and real estate are skewed by their utility value. And bonds are held for all sorts of weird reasons -- everything from governments trying to control the strength of their currency to giant mortgage companies hedging their loans.
I use bonds to dampen portfolio volatility and provide guaranteed income, so I tune my allocation to suit my needs (and volatility fears).
I use real assets to ensure that I'm always holding something that will continue to have fungibility in the face of just about any economic storm. So, I adjust my allocation based on my worst-case predictions for other assets.
I guess I take some solace in the lack of historical covariance among these assets too, but I don't really use that to help with allocation decisions.
I like the idea Sharpe hits upon - you've got to start somewhere - and he proposes that it makes sense to start with a world portfolio as a basis. In other words, start with the concept of world cap-weighted portfolio and then start chipping away and making changes to fits one's needs. That makes sense to me. Why start with a cap-weighted world portfolio as a basis? Because any time I stray from the market as a whole I'm engaging in a zero sum game - for every winner there will be a loser relative to the entire market - I'm placing "bets". So every diversion should be made for a good reason. I'm still struggling with my own allocation. My total market approach has prevailed - but I may make a couple of side bets:
1. I'm sticking with my home bias because my risk tolerance just won't handle ~60% international.
2. I may tilt toward value stocks because I prefer the higher income, and I'm inclined to think value may provide a higher return going forward.
3. A dose of REITs makes sense to me.
But my bets will be relatively small. And I'm talking about stocks only - I look at bonds entirely differently for many of the reasons Wab mentions above. When it comes to bonds, I'll buy whatever I believe will provide a reasonably safe income stream, and a cash flow that fits my needs.
Wab,
Interesting thought experiment, but two points I'd add to counter (I am a splitter, btw).
a) you only work with real asset classes that have meaningful long term track records. (Your lemonade stand won't qualify!). Small Cap in most of these definitions, (e.g. DFA's) is the bottome 10 or 20% of NYSE stocks. These are what any one of us would recognize as massive, successful companies.
The other point is that Small Cap Value have trounced everything over the last 75 years -- people don't want to hold things that are undervalued (down on their luck?) or small, so these companies are underpriced, as an asset class. (They fix their problems and grow, which makes their prices go up eventually).
DFA's data book shows long term rates (1927-2002) at 14.2% vs 9.4% for Large Cap Growth over the same period, or 10.2 for the S&P 500.
Slicers get access to these less-correlated asset classes with less volatility of the SP500. (The less correlated asset classes being included tends to damp volatility - Modern Portfolio Theory).
We had a thread six months ago or so that talked about how the SP500 is actually a huge hype and is not the index that we want to follow -- it has huge concentration in its megacap stocks -- as I recall, 35% of the index was tied up in just 15 or so stocks, and the smallest 150 stocks made up a few percent of the index.
Not anyone's idea of diversification -- rather something that has been sold to us imho!
ESRBob
wabmester
Thinks s/he gets paid by the post
- Joined
- Dec 6, 2003
- Messages
- 4,459
Bob, all I'm saying is that if you believe the markets are efficient, then investors have already considered all available information -- including the historical data you and I have seen -- and they have made their bets. The result is the market cap of each company in the stock market. Everybody is free to allocate anyway they want to, but overweighting in small/value is essentially a bet against market efficiency and an implicit statement that you're smarter than the market.
Now, what if I told you that over the last 75 years, companies whose names began with the letter "M" have outperformed the market as a whole and with less volatility to boot.
I've heard all of the rationalizations for why small/value (for a given arbitrary definition of the class) outperforms the market. But I'm not sure the predictive value of that historical data is any better than the "M" company data-mining, but who knows....
Now, what if I told you that over the last 75 years, companies whose names began with the letter "M" have outperformed the market as a whole and with less volatility to boot.
I've heard all of the rationalizations for why small/value (for a given arbitrary definition of the class) outperforms the market. But I'm not sure the predictive value of that historical data is any better than the "M" company data-mining, but who knows....
Similar threads
Latest posts
-
-
-
-
-
In Your City - What Are The Most Desirable Neighborhoods and Why?
- Latest: RetiredAndFree
-
Best CD, MM Rates & Bank Special Deals Thread 2024 - Please post updates here
- Latest: Healthy Lifestyle
-
