Article suggests "differentiated" asset locations superior

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.

Article suggests "differentiated" asset locations superior

Postby tj-longterm » Tue Jul 16, 2013 7:28 am

For investors with primarily taxable, not tax-sheltered investments, a new article suggests the differentiated asset location is better than segregating tax-inefficient asset classes into retirement accounts:

https://blog.wealthfront.com/differenti ... investors/

I'd love to get everyone's opinions on this:

Differentiated asset location differs from segregated asset location in that it focuses on calculating allocations that deliver the best after-tax returns for each type of account – be it retirement or taxable. To make decisions about differentiated asset location, we consider the tax characteristics of an asset class, its potential for risk-adjusted return, and the way it balances the risks of the other asset classes in the account. In other words the allocation is not solely based on tax efficiency or inefficiency.


It's a little unclear how to apply this, and I suppose their answer would be "let us manage your money for you and we'll do it", but I wonder if there is anything to their arguments, which differ from the standard Bogleheads approach to asset location, and how to replicate it in your own portfolio if so.
tj-longterm
 
Posts: 479
Joined: Tue Apr 10, 2007 10:10 am

Re: Article suggests "differentiated" asset locations superi

Postby dhodson » Tue Jul 16, 2013 7:37 am

Id bet you a virtual beer that their idea involves permanent insurance.
dhodson
 
Posts: 3503
Joined: Mon May 24, 2010 3:03 pm

Re: Article suggests "differentiated" asset locations superi

Postby tj-longterm » Tue Jul 16, 2013 8:50 am

dhodson wrote:Id bet you a virtual beer that their idea involves permanent insurance.


Not sure what you mean -- they give an example of a differentiated portfolio in the blog post, and specifically recommend against whole life. It's simply a different method of allocating assets between retirement and taxable accounts. Instead of putting the most tax-inefficient assets in retirement accounts first, they seem to have a different formula to maximize the after-tax returns.

This isn't some random high expense financial firm, their investment team is run by Burton Malkiel (of A Random Walk Down Wall Street fame) and are fee-only (0.25%/year with the first $10,000 free). They recommend almost exclusively Vanguard ETF indexes, with a couple of iShares options.

But, has this "differentiated" asset location approach been discussed anywhere else? All I can find is their white paper that suggests they do a mean-variance optimization to optimize the returns:

https://www.wealthfront.com/whitepapers ... ethodology

I think you owe me a virtual beer :sharebeer

Edit: Just to be clear, I'm not recommending Wealthfront. Some previous discussions on here highlight some warning flags and other issues that wouldn't make them a great idea over either doing it yourself or hiring a good fee-only advisor. But, I am curious about this alternate approach to allocating assets between taxable and retirement accounts and whether or not it's worth considering.
Last edited by tj-longterm on Tue Jul 16, 2013 9:13 am, edited 2 times in total.
tj-longterm
 
Posts: 479
Joined: Tue Apr 10, 2007 10:10 am

Re: Article suggests "differentiated" asset locations superi

Postby dhodson » Tue Jul 16, 2013 9:11 am

tj-longterm wrote:
dhodson wrote:Id bet you a virtual beer that their idea involves permanent insurance.


Not sure what you mean -- they give an example of a differentiated portfolio in the blog post, and specifically recommend against whole life. It's simply a different method of allocating assets between retirement and taxable accounts. Instead of putting the most tax-inefficient assets in retirement accounts first, they seem to have a different formula to maximize the after-tax returns.

This isn't some random high expense financial firm, their investment team is run by Burton Malkiel (of A Random Walk Down Wall Street fame) and are fee-only (0.25%/year with the first $10,000 free). They recommend almost exclusively Vanguard ETF indexes, with a couple of iShares options.

But, has this "differentiated" asset location approach been discussed anywhere else? All I can find is their white paper that suggests they do a mean-variance optimization to optimize the returns:

https://www.wealthfront.com/whitepapers ... ethodology

I think you owe me a virtual beer :sharebeer


you are probably right. Drink up.
Some of the wording made me suspicious of VULs as well as the lack of a clear explanation in my view of "differentiated". Ill have to look at that white paper when i get a chance.
dhodson
 
Posts: 3503
Joined: Mon May 24, 2010 3:03 pm

Garbage In/Garbage Out

Postby House Blend » Tue Jul 16, 2013 9:21 am

Garbage In/Garbage Out.

They make a bunch of assumptions about expected returns and tax costs
of 9 asset classes. Then "We feed these assumptions into a mean variance optimization framework to calculate optimal asset allocations for various mixes".

Sigh. This is magical thinking. There's nothing remotely resembling signal in there amongst the noise that would support meaningful MVO for as few as 3 asset classes. Probably not even two.

I love this footnote: "Please note this is just one example. If you use different assumptions you will get different results."

Hucksters trying to suck in the gullible.

Update: Just noticed that they don't even mention the assumptions they make about the variance and correlations of these 9 asset classes. Can't do MVO without it.
User avatar
House Blend
 
Posts: 2487
Joined: Fri May 04, 2007 1:02 pm

Re: Article suggests "differentiated" asset locations superi

Postby gnav » Tue Jul 16, 2013 9:46 am

"Differentiated Asset Location" appears to be a fancy term for what firms such as Portfolio Solutions already do, allocate a mix of equities and bonds into both retirement and non-retirement accounts. The potential boost to total return appears to be minimal. Benefits are largely behavioral, accounts are never greatly out of sync with each other up or down. And this makes it somewhat easier to rebalance.
gnav
 
Posts: 13
Joined: Mon Jun 24, 2013 11:44 pm

Re: Article suggests "differentiated" asset locations superi

Postby gnav » Tue Jul 16, 2013 1:50 pm

Here's an article which argues for placing equities in tax-deferred and roth accounts.
http://tinyurl.com/krvoysd
It's an interesting debate, see the comments.

For simplicity's sake, I favor the Bogleheads recommendation for asset location, tax-inefficient in retirement accounts, tax-efficient in taxable.
But I can also appreciate the psychological benefits of Rick Ferri's approach, balance your stock and bond holdings across all accounts.
I'd say pick one or the other. Anything more complicated is likely to generate more noise than substance in the long run.
gnav
 
Posts: 13
Joined: Mon Jun 24, 2013 11:44 pm

Re: Article suggests "differentiated" asset locations superi

Postby grabiner » Wed Jul 17, 2013 7:58 pm

I think the advantage of their differentiated asset location is not the result of differentiation, but of the conventional wisdom being inconsistent with their assumptions. If corporate bonds yield 1.6% and municipal bonds yield 1.1%, then the tax cost of holding bonds in a taxable account is 0.5%, which is less than for any other asset class; therefore, the conventional wisdom of holding bonds in a tax-sheltered account leads to an inferior asset location. Conversely, the "dividend stocks" are assumed to have half their return from dividends; such stocks should be in tax-deferred even in an index fund.

But those bond yields are unrealistic unless the funds are high-expense. Admiral shares of Vanguard Intermediate-Term Tax-Exempt yield 2.4%, and my 25% break-even rule of thumb says that a corporate fund of comparable risk would yield 3.2%.

In addition, I believe the article assumes a very high tax bracket, as it gives a 0.8% after-tax return for corporate bonds with a 1.6% pre-tax return; presumably, this is the top 43.4% federal tax bracket (and 23.6% on qualified dividends), which makes stocks in taxable even more costly.
David Grabiner
User avatar
grabiner
Advisory Board
 
Posts: 13696
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD


Return to Investing - Theory, News & General

Who is online

Users browsing this forum: chaz, columbia, Flobes, furwut, Ged, Google [Bot], Majestic-12 [Bot], packet, tomd37, Yahoo [Bot] and 87 guests