Logical Asset Allocation

[scojo]

Recently, on another thread, there was some interesting discussion regarding a logical investment strategy.  Words I don't understand like "a priori" and "a posteriori" were thrown around.  Sol had this to say on the matter:

my logic for choosing passive index investing is that it is guaranteed to track the market at the lowest possible cost.  Not outperform. Nothing to do with past performance. Not based on any preconceived notion of market behavior.  Just capture everything the market does, as a mathematical certainty, at minimum cost.

Now, since I am currently in the process of trying to determine my own Asset Allocation, it led me to the question: Can an asset allocation be determined "a priori" in the same way that the Passive Indexing Buy and Hold investment strategy claims to be?

Focusing on stocks, the only allocation that I can think of that follows the logic presented by Sol is 100% Total World Stock Index. 

So, in this thread, I would love to hear people's justifications for a U.S. Tilt. Or a small cap Tilt.  Or a Value Tilt.

The part that I guess I am struggling with is where you draw the line in the sand when determining how much to tilt.  Vanguard seems to recommend 20%-40% international, even though international equities make up 49% of VTWSX.

How do you justify overweighting U.S. Stocks, and how do you determine how much to do so?

Hopefully I'm expressing my thoughts in an understandable manner.  I tend to go in circles a bit when I try to think about these things...

[forummm]

Recently, on another thread, there was some interesting discussion regarding a logical investment strategy.  Words I don't understand like "a priori" and "a posteriori" were thrown around.  Sol had this to say on the matter:

my logic for choosing passive index investing is that it is guaranteed to track the market at the lowest possible cost.  Not outperform. Nothing to do with past performance. Not based on any preconceived notion of market behavior.  Just capture everything the market does, as a mathematical certainty, at minimum cost.

Now, since I am currently in the process of trying to determine my own Asset Allocation, it led me to the question: Can an asset allocation be determined "a priori" in the same way that the Passive Indexing Buy and Hold investment strategy claims to be?

Focusing on stocks, the only allocation that I can think of that follows the logic presented by Sol is 100% Total World Stock Index. 

So, in this thread, I would love to hear people's justifications for a U.S. Tilt. Or a small cap Tilt.  Or a Value Tilt.

The part that I guess I am struggling with is where you draw the line in the sand when determining how much to tilt.  Vanguard seems to recommend 20%-40% international, even though international equities make up 49% of VTWSX.

How do you justify overweighting U.S. Stocks, and how do you determine how much to do so?

Hopefully I'm expressing my thoughts in an understandable manner.  I tend to go in circles a bit when I try to think about these things...

People choose to overweight various things for various reasons. Overweighting US is frequently justified by US citizens with reasons such as it minimizes currency risk, it's slightly more tax optimized, and there are slightly lower fees. If you're going to overweight something, you'll have to decide what percent to do it and why. There's no magic number--you're making it up by deviating from the market signal.

[MDM]

Recently, on another thread, there was some interesting discussion regarding a logical investment strategy.  Words I don't understand like "a priori" and "a posteriori" were thrown around.

You could google either of those phrases.

Can an asset allocation be determined "a priori" in the same way that the Passive Indexing Buy and Hold investment strategy claims to be?

Sure, an asset allocation can be determined.  But if you are asking if the best asset allocation can be determined, the answer is "no."

See this chart (there are similar ones at other sites) for year to year changes in "the best":

[CanuckExpat]

You could weight your asset allocation by the individual size of the bond and equity markets if you wanted. You would hold much less equities than most people trying to retire early. The bond market is about 2x as big as the global equity markets, depending on when you count.
https://en.wikipedia.org/wiki/Stock_market#Size_of_the_market
https://en.wikipedia.org/wiki/Bond_market#Bond_market_size
http://finance.zacks.com/bond-market-size-vs-stock-market-size-5863.html

So if you wanted to pick that as your yard stick, you'd be somewhere around 60/40 bonds/equities. It'd be a reasonable allocation for a lot of things: relatively stability, some equity growth, income. In fact, it's probably the large number of institutional investors looking to park some money that the bond market is so huge. The question is, are their needs the same as yours?

If those meet your needs, or you'd prefer to be more weighted in the equity market is up to you.

[Indexer]

Plenty of good advice already listed but I did want to comment on one thing.

Focusing on stocks, the only allocation that I can think of that follows the logic presented by Sol is 100% Total World Stock Index

If you did the total stock and the total international stock you can get the total world, but at a much lower expense.  There is no reason to do the total world. 

[forummm]

Plenty of good advice already listed but I did want to comment on one thing.

Focusing on stocks, the only allocation that I can think of that follows the logic presented by Sol is 100% Total World Stock Index

If you did the total stock and the total international stock you can get the total world, but at a much lower expense.  There is no reason to do the total world. 

50% VTSAX and 50% VTIAX approximates the total world.

[sol]

Oh man, if people are basing their investment strategies off of some BS I was randomly shouting at the internet last week, we know we're in trouble.

Scojo, most people here would recommend that your asset allocation be determined by your personal goals and risk tolerance.  There are valid reasons to deviate from whole-world index exposure.  Let me think of a few...

1.  total portfolio size.  If you already have a billion dollars, and have low expenses, you might consider more conservative fixed income investments because capital preservation is more important to you than growth.
2.  outside assets, like a pension, that already provide the equivalent of fixed income investments and thus might suggest you should lever up the equity exposure in the rest of your portfolio.
3.  home country bias, if you're trying to minimize currency risk or expense ratios.
4.  tax shelter availability, if the benefits of the sheltering specific asset types outweigh the potential losses associated with an unbalanced portfolio.
5.  time horizon or liquidity issues.  If you expect to need the money in six months, the 30 year treasury market isn't relevant to you.
6.  lofty goals.  If you are young and ambitious and poor, you might accept crazy high risks for crazy high rewards because your wipeout risk is minimally dangerous to your long term plans.
7.  insider information.  People fled Pyongyang before the Korean War started because they could read the writing on the wall, and in some cases maybe you can too.
8.  familiarity.  I wouldn't invest in anything I didn't at least sort of understand, regardless of it's market cap.
9.  the realization that market cap is totally bogus anyway.  The value of all financial assets in the world is roughly equal to the value of all real estate in the world, but very few people take this to mean they should have 1/2 of their total investments in real estate.  The total value of derivatives is WAY higher than the total value of all financial assets, but you'd be stupid to put all of your money in derivatives. You have to limit your definition of "whole world" to some subset of asset classes, and how you do that is kind of an open question. 

Then there's the issue of whether or not indexing even makes sense.  There's a whole sub-cult on this forum (see miles, hode) who believe that your investment choices should follow strategies that have done well in the past, that you can do better than an index in the future if you invest in ways that have done better than an index in the past.  This is purely a philosophical leap of faith, but it's a pretty common one.  And only time will tell.

[NorCal]

Asset allocation has to be based on your individual risk/return needs & tolerance.

Here is the best information to make your decision between US vs. International.  Note that "International" can be weighted many different ways which would influence the chart.

http://www.bogleheads.org/wiki/File:US-International.png

[innerscorecard]

There is no a-priori "best," and there are always tradeoffs. Instead of trying to find the optimal, most people should just find something simple enough to repeat to a fifth-grader and stick with it.

For example, there's a lot you can find wrong with 1/3 VTI, 1/3 VXUS and 1/3 BND. But it's easy to understand, and if you stick with it, you'll do ok given a long enough time horizon (which most people don't have, even if they think they do, but that's a matter for another day).

[arebelspy]

There is no a-priori "best," and there are always tradeoffs.

+1.

It should be tailored to you.

[milesdividendmd]

Oh man, if people are basing their investment strategies off of some BS I was randomly shouting at the internet last week, we know we're in trouble.

Scojo, most people here would recommend that your asset allocation be determined by your personal goals and risk tolerance.  There are valid reasons to deviate from whole-world index exposure.  Let me think of a few...

1.  total portfolio size.  If you already have a billion dollars, and have low expenses, you might consider more conservative fixed income investments because capital preservation is more important to you than growth.
2.  outside assets, like a pension, that already provide the equivalent of fixed income investments and thus might suggest you should lever up the equity exposure in the rest of your portfolio.
3.  home country bias, if you're trying to minimize currency risk or expense ratios.
4.  tax shelter availability, if the benefits of the sheltering specific asset types outweigh the potential losses associated with an unbalanced portfolio.
5.  time horizon or liquidity issues.  If you expect to need the money in six months, the 30 year treasury market isn't relevant to you.
6.  lofty goals.  If you are young and ambitious and poor, you might accept crazy high risks for crazy high rewards because your wipeout risk is minimally dangerous to your long term plans.
7.  insider information.  People fled Pyongyang before the Korean War started because they could read the writing on the wall, and in some cases maybe you can too.
8.  familiarity.  I wouldn't invest in anything I didn't at least sort of understand, regardless of it's market cap.
9.  the realization that market cap is totally bogus anyway.  The value of all financial assets in the world is roughly equal to the value of all real estate in the world, but very few people take this to mean they should have 1/2 of their total investments in real estate.  The total value of derivatives is WAY higher than the total value of all financial assets, but you'd be stupid to put all of your money in derivatives. You have to limit your definition of "whole world" to some subset of asset classes, and how you do that is kind of an open question. 

Then there's the issue of whether or not indexing even makes sense.  There's a whole sub-cult on this forum (see miles, hode) who believe that your investment choices should follow strategies that have done well in the past, that you can do better than an index in the future if you invest in ways that have done better than an index in the past.  This is purely a philosophical leap of faith, but it's a pretty common one.  And only time will tell.

Sol,

At the risk of being accused of being overly sensitive...

I've got to argue with the whole "sub-cult" status ascribed to my personal use of dual momentum.

I have chosen my specific strategy not simply because it has done well in the past, but primarily because I believe it is likely to truncate drawdowns in the the future.

This trend following approach allows me to comfortably take on more risk than I could in the past (I used to own, 25-30% bonds, currently I own less than 3% and all in my taxable accounts.)

I have no clue whether DM will continue to outperform buying and holding a 100% stock portfolio in the future. But I feel it's very likely that it will outperform my old 70/30 portfolio, and with less drawdowns. More importantly, DM is behaviorally comfortable for me, and very cheap.

But when it comes to advising others on their investment choices I still generally recommend:

1.   Buy and hold low cost index funds
2.   50% international allocation at least.
3.   5 to 10% REIT  exposure
4.   More bonds than you think you need.
5.   Tilting towards small size and value, if that appeals to you. (It appeals to me.)

(Happy to defend any of my recs that anyone disagrees with.)

So that would be what I would argue for the OP to consider, before ever recommending DM.

AZ

[sol]

I've got to argue with the whole "sub-cult" status ascribed to my personal use of dual momentum.

I wasn't trying to be derogatory.  If MMM is a cult, then MMM market timers are a sub-cult.  A minority within the minority.

There are lots of different investment theories presented here, by lots of different people.  I only referenced the Dual Momentum folks because they've been recently popular, numerous, and well defended/explained for anyone who wants to explore.

[milesdividendmd]

Fair enough.

As long as we all agree that I lack the necessary charisma to be labeled a cult leader.

The main points that I was disagreeing with you about in your characterization of my philosophy...

1. While I am eager to defend my own choices I am hesitant to suggest them to others, particularly when they are as niche as DM. I generally recommend buy and hold as the most defensible and smart option.

and.

2.  The chief attraction of DM for me personally is diminished drawdowns, not market beating performance.

I will be truly disappointed if I don't sit out of a big part of the next bear markets, but I won't be terribly bothered if the S&P beats me by a point or 2 of CAGR in the long haul.

Small potatoes. 

[scojo]

Thanks All,

I appreciate the responses and discussion. 

Perhaps I'll just use a die to to determine my International stock slice.  10*(Roll - 1) can be the percentage of the portfolio dedicated to international stocks.  It seems like as good a method as any I can think of for making it "personal".  And this way, if it all goes to shit, I can blame in on the die instead of taking any personal responsibility.

Or, perhaps I'll know my desired allocation before the die stops rolling...

As a side note, asset allocation may have been the wrong choice for the title of the thread.  I'm mostly slicing (or not slicing) up the stock portion of the portfolio, if you hadn't already noticed.