VTIAX or VTMGX and VEMAX

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

VTIAX or VTMGX and VEMAX

Post by Osp62 » Fri Apr 17, 2015 3:35 pm

VTIAX total international - 0.14% fee, VTMGX: Dev Intl.- 0.09% fee, VEMAX- emerging market - fee:0.15%. Total International is very similar to Dev international and emerging market in the ratio of 4:1. If you held VTMGX and Vemax in the ratio 4:1 your fees would be close to .102%. Wondering if it makes sense to pay additional 3.8 basis points in fees for total international.

I understand that total international is a larger fund, it includes Canada and has a lot more stocks than developed international +emerging market funds and will automatically rebalance between developed and emerging markets. How do you put a value in these additional things that total international provides and can you say with any certainty that these benefits are worth the additional 3.8 basis points in fees.

Also wondering if anyone has any thoughts on how to estimate tax savings in the two scenarios assuming that VTIAX will most likely be better at managing taxes from rebalancing than an individual investor rebalancing developed international and emerging market funds on an annual or so basis. Thanks.

User avatar
iceport
Posts: 4151
Joined: Sat Apr 07, 2007 4:29 pm

Re: VTIAX or VTMGX and VEMAX

Post by iceport » Fri Apr 17, 2015 4:40 pm

Here are a few thoughts:

1) The difference in ERs is just not very significant. Therefore, there should better reasons to slice and dice international, such as a desire to overweight a certain portion (or portions) of the total international market, or to hope for some sort of rebalancing bonus by maintaining static allocations to certain portions and not letting them float with the market valuations. Or there might be an asset location advantage to splitting the international market, depending on the fund choices in your defined contribution plans.

2) The total international fund includes a small allocation to small caps, whereas the VTMGX/VEMAX combination is virtually devoid of small caps. If you want exposure to small caps, you'd need to add a third fund, like VSS (ER=0.19%).

3) Rebalancing by selling shares in a taxable account should be a rare occurrence. I haven't done it in the last seven years with two taxable positions, and my rebalancing bands have only been exceeded slightly for a week or two once. You can opt out of automatic dividend reinvestment, and just add to any low positions with new purchases, if you're still accumulating.

4) I'd defer to someone like David Grabiner for a tax comparison of the options you presented — apart from rebalancing sales, which could hopefully be avoided — but I doubt there's a huge difference.
"Discipline matters more than allocation.” ─William Bernstein

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Fri Apr 17, 2015 6:49 pm

Thanks Iceport. This is indeed for a taxable account. I am also not too keen to have small cap exposure in international stocks. I am more concerned about the governance of international small caps (especially small cap emerging market) than about the opportunity cost of missing international small caps. I am also fine not having Canada exposure as I think it is quite correlated with the rest of developed international anyway.

The main issue to me is whether the 4 bps fees extra fee is worth the possible higher long term return and possible tax savings from the automated rebalancing done by vtiax between Developed and emerging markets. I am assuming that If I did the rebalancing between emerging market and developed market myself, I would be much lazier than vtiax and would also not be as tax efficient because I would just do it once or twice a year and let the asset allocation drift more than is optimal.

User avatar
iceport
Posts: 4151
Joined: Sat Apr 07, 2007 4:29 pm

Re: VTIAX or VTMGX and VEMAX

Post by iceport » Sat Apr 18, 2015 10:40 am

Okay, lets focus on rebalancing. You speak of the total international index fund "automatically rebalancing" between developed and emerging markets, as if there's a fixed proportion between the two that the fund maintains. That's not my understanding of how it works. Depending on shifting relative valuations, the market share split between developed and emerging markets changes over time. When emerging market valuations are high relative to developed markets, the total international index will hold a higher percentage of emerging markets, and vice versa. The index knows nothing of an "optimum" proportion between the two portions of the market, but rather just lets it float with the market cap weightings. So in essence, the total international index fund requires no rebalancing between developed and international.

In this respect, the total international index fund is like the total US market index fund. The higher the value of Apple stock, the higher the proportion of Apple stock in the total US market index. If it rockets up 25% in a few months, the index simply lets the percentage of Apple stock float up right along with it. That's my understanding of how the total international index works: it holds the market cap weighting of developed and emerging markets. As that market cap weighting shifts over time, the index lets the proportion float along with it. (This is distinctly different than, say, a target retirement fund that wants to hold a designated percentage of US equity, international equity, and fixed income. In this scenario, the fund must make frequent adjustments to maintain the established targets.)

Now, let's say you choose to split developed and emerging market allocations. Usually, this is done by picking a proportion between the two and maintaining it over time. I've never read others describe monitoring the relative proportion between developed and emerging markets and attempting to match it with rebalancing, though I suppose someone could do that. (The thing is, they probably wouldn't need much rebalancing.)

If you chose a fixed proportion of, say, 20% emerging/80% developed, then over time the markets will drift out of balance. A good year for emerging markets? Chances are you'll want to add to the developed markets fund. By doing that, you are essentially topping up the allocation of a fund with a lower valuation — and possibly higher expected future return. That's the way the theoretical "rebalancing bonus" could work. So while there's no guarantee you'd be fortunate in the timing of your rebalancing, you could theoretically achieve better returns by holding separate, fixed percentages of emerging and developed markets than by holding a total international fund that lets the proportion float.

Would it work out that way for you? Who knows? But it is more work. Usually, people split the markets if they want to overweight one portion or the other, or if they are missing something in their 401k choices.

And I'll reiterate that your goal in a taxable account should be to avoid selling anything for the sole purpose of rebalancing. So any difference in tax efficiency should just be related to the differences in tax efficiency between the funds under consideration. Theoretically, a total market fund should be more tax efficient. In practice, I doubt the difference is very big.
"Discipline matters more than allocation.” ─William Bernstein

livesoft
Posts: 69606
Joined: Thu Mar 01, 2007 8:00 pm

Re: VTIAX or VTMGX and VEMAX

Post by livesoft » Sat Apr 18, 2015 11:03 am

Tax efficiency is something that is kind of quirky. I have calculated the tax costs for someone in the 25% marginal income tax bracket for a number of international funds. See viewtopic.php?f=10&t=158519

Here is a line-up for 2014:
Tax-cost Ticker
0.245% SCZ
0.255% FSGDX
0.293% VTI (total US)
0.340% VSS
0.342% VBR (US small-cap value)
0.359% VWO
0.409% VXUS
0.430% VEU
0.457% VEA

While FSGDX might look good, it trails performance-wise (total return) behind VXUS, so one will end up with more after-tax with VXUS.

Furthermore, I think the rank order probably changes every year. And don't forget that in 2014, some international funds lost money, so they were very tax efficient for those folks that did tax-loss harvesting.

Bottom line: I would not give much weight to tax-efficiency when making a decision on this. Convenience and tax-loss harvesting would have much more weight.
Wiki This signature message sponsored by sscritic: Learn to fish.

alex_686
Posts: 5154
Joined: Mon Feb 09, 2015 2:39 pm

Re: VTIAX or VTMGX and VEMAX

Post by alex_686 » Sat Apr 18, 2015 11:07 am

iceport wrote:Now, let's say you choose to split developed and emerging market allocations. Usually, this is done by picking a proportion between the two and maintaining it over time. I've never read others describe monitoring the relative proportion between developed and emerging markets and attempting to match it with rebalancing, though I suppose someone could do that. (The thing is, they probably wouldn't need much rebalancing.)
There would be some need for rebalancing. Over time the balance of these 2 funds would drift a little. As a counterexample, assume one had a large cap value and a large cap growth fund with a initial split of 50/50. As time passes one market would do better than the other, stocks would be moved from one index to the next. etc.

ON a different note, I would hold the 2 funds and I would not hold them in portion to the world market. As equities go they have a low correlation and are driven by different factors. In short, a different risk / return profile. Personally, I modestly overweight my EM, but I am weighting my portfolio my to a higher risk / return profile.

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Sat Apr 18, 2015 4:57 pm

iceport wrote:Okay, lets focus on rebalancing. You speak of the total international index fund "automatically rebalancing" between developed and emerging markets, as if there's a fixed proportion between the two that the fund maintains. That's not my understanding of how it works. Depending on shifting relative valuations, the market share split between developed and emerging markets changes over time. When emerging market valuations are high relative to developed markets, the total international index will hold a higher percentage of emerging markets, and vice versa. The index knows nothing of an "optimum" proportion between the two portions of the market, but rather just lets it float with the market cap weightings. So in essence, the total international index fund requires no rebalancing between developed and international.

In this respect, the total international index fund is like the total US market index fund. The higher the value of Apple stock, the higher the proportion of Apple stock in the total US market index. If it rockets up 25% in a few months, the index simply lets the percentage of Apple stock float up right along with it. That's my understanding of how the total international index works: it holds the market cap weighting of developed and emerging markets. As that market cap weighting shifts over time, the index lets the proportion float along with it. (This is distinctly different than, say, a target retirement fund that wants to hold a designated percentage of US equity, international equity, and fixed income. In this scenario, the fund must make frequent adjustments to maintain the established targets.)

Now, let's say you choose to split developed and emerging market allocations. Usually, this is done by picking a proportion between the two and maintaining it over time. I've never read others describe monitoring the relative proportion between developed and emerging markets and attempting to match it with rebalancing, though I suppose someone could do that. (The thing is, they probably wouldn't need much rebalancing.)

If you chose a fixed proportion of, say, 20% emerging/80% developed, then over time the markets will drift out of balance. A good year for emerging markets? Chances are you'll want to add to the developed markets fund. By doing that, you are essentially topping up the allocation of a fund with a lower valuation — and possibly higher expected future return. That's the way the theoretical "rebalancing bonus" could work. So while there's no guarantee you'd be fortunate in the timing of your rebalancing, you could theoretically achieve better returns by holding separate, fixed percentages of emerging and developed markets than by holding a total international fund that lets the proportion float.

Would it work out that way for you? Who knows? But it is more work. Usually, people split the markets if they want to overweight one portion or the other, or if they are missing something in their 401k choices.

And I'll reiterate that your goal in a taxable account should be to avoid selling anything for the sole purpose of rebalancing. So any difference in tax efficiency should just be related to the differences in tax efficiency between the funds under consideration. Theoretically, a total market fund should be more tax efficient. In practice, I doubt the difference is very big.
Thank you Iceport. I understand this better now. I just need to buy in a ratio using today's relative market cap weights and then the two will most likely maintain the relative market cap weights automatically. I am however not sure I follow why I would want to top up the one that has lower growth. If dev markets has slower growth then it just means that now it constitutes a smaller percentage of total international (say has dropped from 80% to 75%) so I can just leave them the way they are and automatically maintain market cap weights. I am not sure of this - perhaps there may be a need for rebalancing as more companies go public in emerging markets and the emerging market market cap based weight is affected by not just its return but also by the additional companies going public.

I am also assuming that to maintain market cap weighted combination you would not want to reinvest dividends automatically. If you want to reinvest dividends you would take the total dividends received and then manually split the total and invest in your funds to rebalance and get closer to current market cap weights.

User avatar
iceport
Posts: 4151
Joined: Sat Apr 07, 2007 4:29 pm

Re: VTIAX or VTMGX and VEMAX

Post by iceport » Sat Apr 18, 2015 5:59 pm

As with many aspects of investing, the implementation is more challenging than the theory. One thing we've been ignoring is that — presuming you have established one — you will still need to maintain some target international allocation. So some rebalancing will always be required. Rebalancing two funds is more complicated when one will do. Rebalancing in a taxable account can be more challenging also, because you want to be careful not to trigger taxable events (by selling with a capital gain). My preference is to hold as few funds as possible in a taxable account. The trade-off is fewer tax-loss harvesting opportunities.

That's really how I would be looking at this if I were you: staying with one total market fund for ease of portfolio management, or splitting them for more tax-loss harvesting opportunities. (Livesoft got there with few words.)

Personally, I'd favor the total market fund, because 1) it's a fund that will always be appropriate to own (so can be held indefinitely); 2) it should be more tax-efficient in the long run than a fund that theoretically requires higher turnover; 3) you prefer the market cap weighting anyway; 4) it's easier to manage; and 5) tax-loss harvesting opportunities eventually dry up anyway. EDIT: and 6) it's better diversified.

The main reasons not to use automatic dividend reinvestment in a taxable account is to reduce the need for rebalancing through a sale, and to prevent a wash sale if a tax-loss harvesting opportunity presents itself. It also reduces the generation of non-qualified dividends.

Whenever you buy a fund in a taxable account, think in terms of being able to hold it "forever" if needed, for tax reasons. You might not need to, but if all goes well, you just might want to. (As an example, the total US stock market index fund I now own was bought near the market bottom of the Great Recession in a tax-loss harvesting move. Roughly half of its current value consists of unrealized capital gains. Obviously, I don't want to sell any of that holding now, only to pay taxes on 50% of the proceeds, if I can avoid it.)
Last edited by iceport on Sun Apr 19, 2015 7:17 am, edited 1 time in total.
"Discipline matters more than allocation.” ─William Bernstein

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Sun Apr 19, 2015 12:05 am

iceport wrote:As with many aspects of investing, the implementation is more challenging than the theory. One thing we've been ignoring is that — presuming you have established one — you will still need to maintain some target international allocation. So some rebalancing will always be required. Rebalancing two funds is more complicated when one will do. Rebalancing in a taxable account can be more challenging also, because you want to be careful not to trigger taxable events (by selling with a capital gain). My preference is to hold as few funds as possible in a taxable account. The trade-off is fewer tax-loss harvesting opportunities.

That's really how I would be looking at this if I were you: staying with one total market fund for ease of portfolio management, or splitting them for more tax-loss harvesting opportunities. (Livesoft got there with few words.)

Personally, I'd favor the total market fund, because 1) it's a fund that will always be appropriate to own (so can be held indefinitely); 2) it should be more tax-efficient in the long run than a fund that theoretically requires higher turnover; 3) you prefer the market cap weighting anyway; 4) it's easier to manage; and 5) tax-loss harvesting opportunities eventually dry up anyway.

The main reasons not to use automatic dividend reinvestment in a taxable account is to reduce the need for rebalancing through a sale, and to prevent a wash sale if a tax-loss harvesting opportunity presents itself. It also reduces the generation of non-qualified dividends.

Whenever you buy a fund in a taxable account, think in terms of being able to hold it "forever" if needed, for tax reasons. You might not need to, but if all goes well, you just might want to. (As an example, the total US stock market index fund I now own was bought near the market bottom of the Great Recession in a tax-loss harvesting move. Roughly half of its current value consists of unrealized capital gains. Obviously, I don't want to sell any of that holding now, only to pay taxes on 50% of the proceeds, if I can avoid it.)
Thanks for your thoughtful response Iceport. Would greatly help if you could also comment specifically on how you think the benefits of total intl. fund justify a 4 basis point higher annual expense for the total international vs. holding a weighted combo of dev intl. and emerging mkt index funds.

User avatar
iceport
Posts: 4151
Joined: Sat Apr 07, 2007 4:29 pm

Re: VTIAX or VTMGX and VEMAX

Post by iceport » Sun Apr 19, 2015 7:56 am

Osp62 wrote:Thanks for your thoughtful response Iceport. Would greatly help if you could also comment specifically on how you think the benefits of total intl. fund justify a 4 basis point higher annual expense for the total international vs. holding a weighted combo of dev intl. and emerging mkt index funds.
Osp62,

I'm not sure they do. We just don't know in advance whether the additional diversification of the total market fund will end up overcoming the 0.038% additional expense over your investing time frame. (Though there are thousands more holdings in the total market fund, they amount to a marginal percentage of the fund.)

But the difference in expense isn't huge, either, So at some point it will come down to how much that means to you, in your personal circumstances. If you're investing in a taxable account, you probably have all available tax-advantaged space used up. So say you have $1M, in 70% equities, and 30% international. The cost of 0.038% on $210k is about $80/year. If you have ten times that in international funds, it will cost you $800/year, and so on. How big a deal is that to you? If you don't mind tracking and rebalancing two taxable positions instead of one, and somewhat less diversification does not concern you, then by all means, split them up. Who knows, you might even enjoy a rebalancing bonus if you establish fixed target percentages and rebalance to them. (Just be sure you don't trigger taxable events that offset the savings!)

Ironically, I'm about to add a third taxable fund to my VTSAX and VEMAX positions: VTMGX. (But I have some excuses for splitting international: my employer's defined contribution plan holds only an EAFE index fund (TCIEX), no emerging markets; I want to overweight emerging markets somewhat, anyway; and I also overweight small caps with VSS.)

Hopefully you'll get some other opinions on this. If not, it might just be because others don't see a huge difference, either.
"Discipline matters more than allocation.” ─William Bernstein

livesoft
Posts: 69606
Joined: Thu Mar 01, 2007 8:00 pm

Re: VTIAX or VTMGX and VEMAX

Post by livesoft » Sun Apr 19, 2015 8:30 am

The extra 4 basis points get you international small caps in both developed and emerging markets for about 10% of the fund value. Definitely worth it. As shown in my post above on tax costs, small caps are more tax efficient than the large caps probably simply from their lower dividends.
Wiki This signature message sponsored by sscritic: Learn to fish.

User avatar
iceport
Posts: 4151
Joined: Sat Apr 07, 2007 4:29 pm

Re: VTIAX or VTMGX and VEMAX

Post by iceport » Sun Apr 19, 2015 10:38 am

livesoft wrote:The extra 4 basis points get you international small caps in both developed and emerging markets for about 10% of the fund value. Definitely worth it.
livesoft,

The OP prefers to avoid international small cap.
Osp62 wrote:I am also not too keen to have small cap exposure in international stocks. I am more concerned about the governance of international small caps (especially small cap emerging market) than about the opportunity cost of missing international small caps.
That's not a risk I've considered in my asset allocation, but there's nothing wrong with consciously avoiding some forms of perceived risk.

Regardless, with respect to tax consequences, expense ratios, marginal small cap exposure, convenience, etc., I think we have found ourselves in that far off place Taylor can spot from a mile away: dancing on the head of a pin!
"Discipline matters more than allocation.” ─William Bernstein

livesoft
Posts: 69606
Joined: Thu Mar 01, 2007 8:00 pm

Re: VTIAX or VTMGX and VEMAX

Post by livesoft » Sun Apr 19, 2015 11:12 am

iceport wrote:The OP prefers to avoid international small cap.
Yes, I was well aware of that when I posted my response. I see no reason to save 4 basis points for a reduced expected return. That is, the expected return from small-caps should overwhelm the 4 basis points.
Wiki This signature message sponsored by sscritic: Learn to fish.

User avatar
Taylor Larimore
Advisory Board
Posts: 28954
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

"Dancing on the head of a pin"

Post by Taylor Larimore » Sun Apr 19, 2015 11:15 am

I think we have found ourselves in that far off place Taylor can spot from a mile away: dancing on the head of a pin!
iceport:

I was thinking exactly those words.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Sun Apr 19, 2015 5:38 pm

Thanks to all for their thoughts. Just realized that I should have done some more homework before asking the question so went to Morningstar and got the past performance data for all the funds and related ETF's. I have also calculated the implied return (pretax and tax adjusted) for an 80-20 weighted portfolio of developed international and emerging markets using VG funds or ETF's. I know past performance is no guarantee for future returns but if you go back long enough they definitely become something we cannot ignore. So here they are - I was definitely surprised. Would be great to hear thoughts others have;

Please Note that some of the data is missing at Morningstar e.g. Morningstar does not have Tax adjusted returns for VTIAX for 5, 10 and 15 yr.

Tax-Adjusted Returns:
(surprisingly VTIAX appears to have done the worst among the three options its returns are lower by 0.86% over the last 3 years):

VTIAX (YTD, 1yr, 3yr, others missing): 3.95, -2.23, 5.35
VTIAX vs VTMGX+VEMAX in 4:1 ratio (YTD, 1yr, 3yr): 0.74, 0.95, 0.86

VTMGX+VEMAX (4:1 ratio) (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 4.69, -1.28, 6.21, 4.51, 3.67, 2.18
EFA+EEM (4:1 ratio) (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr missing): 4.28, -1.82, 6.31, 4.41, 4.85

VTMGX (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 5.36, -2.08, 7.85, 5.37, 4.59, 2.72
EFA (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 4.85, -2.04, 8.12, 5.40, 4.24
VEMAX (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 2.02, 1.92, -0.37, 1.06
EEM (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 2.01 -0.92, -0.91, 0.47, 7.29
VWO (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 2.00, 1.89, -0.38, 1.06, 7.42

Pre-Tax Returns:
(here the data is more complete and again VTIAX seems to be doing the worse by 1.59% over the last 15 years):
VTIAX (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 4.09, -1.05, 6.62, 4.88, 5.24, 2.26
VTIAX vs VTMGX+VEMAX (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 0.78, 0.95, 0.64, 0.43, 0.48, 1.59

VTMGX+VEMAX (4:1 ratio) (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 4.87, -0.10, 7.26, 5.31, 5.72, 3.85
EFA+EEM (4:1 ratio) (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr missing): 4.28, -0.86, 7.08, 5.03, 5.42

VTMGX (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 5.56, -0.89, 8.93, 6.18, 5.11, 2.95
EFA (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr missing): 4.85, -1.05, 8.92, 6.05, 4.84
VEMAX (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 2.10, 3.08, 0.57, 1.82, 8.15, 7.47
EEM (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr missing): 2.01, -0.09, -0.30, 0.95, 7.73
VWO (YTD, 1yr, 3yr, 5yr, 10 yr, 15 yr): 2.08, 3.05, 0.57, 1.82, 8.16

Please let me know if I am missing something. I am definitely surprised to see this because I asked the question about a 4 basis point cost difference and here I see an after tax return difference of 0.86% over 3 years and a Pre-tax difference of 1.59% over 15 years. I am assuming that I must be missing something because I cannot reconcile this with the dramatic growth and current scale of VTIAX. As a note all I did to get the Implied return of the combination of VTMGX and VEMAX was to calculate a weighted average of the two component returns weighted 80% and 20%.

livesoft
Posts: 69606
Joined: Thu Mar 01, 2007 8:00 pm

Re: VTIAX or VTMGX and VEMAX

Post by livesoft » Sun Apr 19, 2015 6:06 pm

Too many numbers for me to try to read and understand, sorry.

I wanted to make the comment that any numbers reported by M* for "tax-cost" are suspect until proven otherwise. I would not use them myself.

Also these funds have been reconstituted a few times over the years, so past performance cannot tell one much of anything about future performance.
Wiki This signature message sponsored by sscritic: Learn to fish.

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Sun Apr 19, 2015 6:14 pm

livesoft wrote:Too many numbers for me to try to read and understand, sorry.

I wanted to make the comment that any numbers reported by M* for "tax-cost" are suspect until proven otherwise. I would not use them myself.

Also these funds have been reconstituted a few times over the years, so past performance cannot tell one much of anything about future performance.
I gave the numbers primarly as backup data. You just need to look at the two bold- faced sentences to see the difference in the tax adjusted and pretax returns between VTIAX and a combination of VTMGX and VEMAX.

User avatar
iceport
Posts: 4151
Joined: Sat Apr 07, 2007 4:29 pm

Re: VTIAX or VTMGX and VEMAX

Post by iceport » Sun Apr 19, 2015 6:49 pm

Osp62, these are index funds. Your primary concerns in selecting index funds should be whether they provide a market exposure consistent with your asset allocation, their costs, and whether they will track their indexes faithfully and efficiently. With Vanguard, you can be reasonably assured of the last point. So the index construction and expenses should be your focus.

Though I recognize your thought process in my own past investing practices, it's been over a decade since I've looked at the past performance of individual funds as a fund selection criteria. In adopting a passive portfolio, my focus has turned to asset allocation, not wasting effort poring over the past performance of individual funds.

Also, livesoft is right, the funds have gone through restructuring, so their past performance may not be representative of their current portfolios anyway.
"Discipline matters more than allocation.” ─William Bernstein

User avatar
grabiner
Advisory Board
Posts: 25652
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: VTIAX or VTMGX and VEMAX

Post by grabiner » Sun Apr 19, 2015 7:12 pm

Osp62 wrote:Please Note that some of the data is missing at Morningstar e.g. Morningstar does not have Tax adjusted returns for VTIAX for 5, 10 and 15 yr.
Don't use Morningstar's tax data; use the data from the fund provider. Morningstar doesn't always know about qualified dividends, and there are a few other errors. For example:

VTMGX: 10-year tax cost 0.49%, 15-year tax-cost 0.22%. This implies a total cost of 4.9% over 10 years and 3.3% over 15 years; that is, from 2000-2004, you got money back in taxes if you held the fund. (Pacific and Emerging Markets have similar problems. This is a long-standing issue; in 2004, the fund reported a negative tax cost, and M* couldn't find the error.)

VEU: 3-year tax cost 1.21%. VFWAX: 3-year tax cost 1.06%. These are the Admiral and ETF classes of the same fund, so they have the same distributions, and the 1-year tax costs are equal. Morningstar presumably is missing some qualified dividend information for VEU. Pacific Index and Value Index have the same problem with 10-year returns.
Wiki David Grabiner

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Sun Apr 19, 2015 10:43 pm

iceport wrote:Osp62, these are index funds. Your primary concerns in selecting index funds should be whether they provide a market exposure consistent with your asset allocation, their costs, and whether they will track their indexes faithfully and efficiently. With Vanguard, you can be reasonably assured of the last point. So the index construction and expenses should be your focus.

Though I recognize your thought process in my own past investing practices, it's been over a decade since I've looked at the past performance of individual funds as a fund selection criteria. In adopting a passive portfolio, my focus has turned to asset allocation, not wasting effort poring over the past performance of individual funds.

Also, livesoft is right, the funds have gone through restructuring, so their past performance may not be representative of their current portfolios anyway.
Thank You Iceport. I agree with your comment about not necessarily looking at historical index fund specific performance and sticking to your asset allocation via index funds. This is why I was initially focused on the fund costs and completely missed the difference in historical returns. Now the real question for me is whether I believe that the market prices for small International companies are efficient enough today to justify a complete bet on VTIAX vs doing the International asset allocation via VTMGX and VEMAX or perhaps even doing 50% of International via VTIAX and 50% as a combo of VTMGX and VEMAX.

If, over the last 15 years, VTIAX has had a 1.59% lower pretax return vs an 80-20 mix of VTMGX and VEMAX it does not seem unreasonable to believe that market pricing for International Larger caps is much more efficient than smaller caps so you may not necessarily be getting adequate return for the risk in the small caps in VTIAX. Even for the US market there may still be some debate (albeit a dying one) about S&P500 index vs Total Market Index. It is noteworthy that even Warren Buffet stated in his 2014 letter that he chose the Vanguard S&P500 index fund for 90% of his wife's inheritance NOT Vanguard Total Market. I continue to be surprised by how his VG Index fund selection got so much press in favor of index funds but there was very little discussion that he explicitly rejected US Total Market in favor of S&P500.

It seems to me that the US financial markets are more efficient than most of the International markets. Just because Total Market Index in the US seems to be performing better than the Large cap Index in the US during the recent past does not necessarily mean that it has to work as well in International markets esp. given how little analyst coverage exists for smaller International companies and how little price discovery happens there (most of the inflow in smaller International markets seems to be index fund or ETF related - so who is setting the prices for these smaller International companies?).

One could also argue that there has been a bandwagon effect that the mainstream media has jumped on regarding TOTAL market weighted index funds so no one is questioning the efficiency of market pricing for any public company. It is not clear to me why the belief in market efficiency for all stock markets in the world has to be absolute. Perhaps it should be more nuanced by at least the company size, analyst coverage, number of savvy active investors helping with price discovery, and especially the quirks of the country market in question. It may be OK to believe in efficient markets and still believe that prices for smaller securities in some International markets may not be efficient and you may not be getting an adequate return for the risk. This would explain the lower historical returns for VTIAX Vs. a combo of VTMGX and VEMAX. It could also say that future returns may continue to be lower or this logic could at least get Investors to question a blind faith in the efficiency of Total International Stock Index Funds.

Thanks also to Grabiner and Livesoft for their comments. I am definitely aware of the inaccuracy of the TaxAdjusted numbers at M* hence I also looked at the PreTax numbers and here the disadvantage for VTIAX is even larger (1.59% over the last 15 years). I am also aware of the changes in the funds esp. the Dev International merger with the tax managed one but the point here is that the return difference is so large (1.59% per year) that it deserves a closer look and the willingness to question the market efficiency of smaller international stocks and hence a complete faith of VTIAX for the International Stock Allocation.

User avatar
ogd
Posts: 4875
Joined: Thu Jun 14, 2012 11:43 pm

Re: VTIAX or VTMGX and VEMAX

Post by ogd » Sun Apr 19, 2015 11:45 pm

Osp62 wrote:If, over the last 15 years, VTIAX has had a 1.59% lower pretax return vs an 80-20 mix of VTMGX and VEMAX it does not seem unreasonable to believe that market pricing for International Larger caps is much more efficient than smaller caps so you may not necessarily be getting adequate return for the risk in the small caps in VTIAX.
No, no, no. This is simply because the proportion 80/20 corresponds to the ratio NOW. It's a rear view mirror effect.

If you backtest sector A and sector B in their relative market cap now, this will ALWAYS beat the combo of sector A and sector B in the combination at the beginning of the period. The difference between the two positions is "+X% of the fund that did best". This happened to be VEMAX (which had a great couple of decades before becoming the dog that it currently is), but going forward it's a big unknown for anyone. Don't read too much else into it.

Also, I find the "blind faith" comment insulting. There are very clear arguments for total market investing. Try to understand them before speculating about why we pick these funds.

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Mon Apr 20, 2015 3:05 am

Thanks OGD. I was not aware of this backward looking point you bring up and it definitely makes sense in general if the combo sector contains nothing else but the two component sectors A and B. In this case however VTIAX contains all that is in VTMGX, all of VEMAX AND a few thousand more small cap stocks.

Note also that 15 year pretax returns of VTIAX (2.26%) are lower than BOTH VTMGX (2.95%) and VEMAX (7.47%). Doesn't this mean that the lower returns are being driven by the constituents of VTIAX that are not part of either VTMGX or VEMAX?

User avatar
iceport
Posts: 4151
Joined: Sat Apr 07, 2007 4:29 pm

Re: VTIAX or VTMGX and VEMAX

Post by iceport » Mon Apr 20, 2015 5:35 am

Osp62 wrote:If, over the last 15 years, VTIAX has had a 1.59% lower pretax return vs an 80-20 mix of VTMGX and VEMAX it does not seem unreasonable to believe that market pricing for International Larger caps is much more efficient than smaller caps so you may not necessarily be getting adequate return for the risk in the small caps in VTIAX.
One flaw in this reasoning is that VTIAX did not have any small caps until December 2010, a little over 4 years ago now.

From the wiki:
On September 24th, 2010, Vanguard announced a change in the index tracked by the Total International fund to the MSCI All Country World ex USA Investable Market Index, and introduced five additional share classes with lower expense ratios for larger holdings. The index change and introduction of admiral shares was completed in December 2010.
I can't seem to find the actual Vanguard announcement easily, sorry. Until the December 2010 implementation of the change in the index it tracks, the fund name was inappropriate, as it was definitely *not* a total market fund in the same way the total stock market fund always was on the domestic side. Be careful about reading too much into past performance, especially when all details are not known.
"Discipline matters more than allocation.” ─William Bernstein

AviN
Posts: 477
Joined: Mon Mar 10, 2014 8:14 am

Re: VTIAX or VTMGX and VEMAX

Post by AviN » Mon Apr 20, 2015 7:09 am

iceport wrote: 3) Rebalancing by selling shares in a taxable account should be a rare occurrence. I haven't done it in the last seven years with two taxable positions, and my rebalancing bands have only been exceeded slightly for a week or two once. You can opt out of automatic dividend reinvestment, and just add to any low positions with new purchases, if you're still accumulating.
If you hold the funds weighted by their market capitalization, you should never need to sell any of the funds to rebalance between them.

User avatar
Doc
Posts: 9307
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: VTIAX or VTMGX and VEMAX

Post by Doc » Mon Apr 20, 2015 7:53 am

livesoft wrote:I wanted to make the comment that any numbers reported by M* for "tax-cost" are suspect until proven otherwise. I would not use them myself.
grabiner wrote:Don't use Morningstar's tax data; use the data from the fund provider. Morningstar doesn't always know about qualified dividends, and there are a few other errors.
Both livesoft and grabiner make good points and I don't disagree that M* data can be flawed. It is only useful for ranking very similar funds if at all. However I also question how you could use actual fund's data. While it may be more accurate one would have to go back and look at several years past data to get an idea of how stable the qualified dividends fraction is. A couple of years of lage short term capital gains distributions might bias the data so much that it not any better than M*'s unless you went to a whole lot of work to refine the numbers and then you still have the assumption of using the past to predict the future.

Another factor that needs to be considered is the "Potential Cap Gains Exposure" which could be important for the future especially if foreign funds are under consideration. This data usually buried in the footnotes in a fund's annual report and can also be found on M*'s tax tab. I assume that this is data has an SEC reporting requirement so either source should be good. How you interpret the number is another matter.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

User avatar
ogd
Posts: 4875
Joined: Thu Jun 14, 2012 11:43 pm

Re: VTIAX or VTMGX and VEMAX

Post by ogd » Mon Apr 20, 2015 1:15 pm

Osp62 wrote:Thanks OGD. I was not aware of this backward looking point you bring up and it definitely makes sense in general if the combo sector contains nothing else but the two component sectors A and B. In this case however VTIAX contains all that is in VTMGX, all of VEMAX AND a few thousand more small cap stocks.

Note also that 15 year pretax returns of VTIAX (2.26%) are lower than BOTH VTMGX (2.95%) and VEMAX (7.47%). Doesn't this mean that the lower returns are being driven by the constituents of VTIAX that are not part of either VTMGX or VEMAX?
Check your numbers again. Performance tab on Morningstar says 4.08% VTIAX vs 3.68 VTMGX. Tax tab, which differs for whatever unknown reason, says 3.26% vs 2.95%.

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Mon Apr 20, 2015 3:14 pm

Thanks OGD. You are right. The 15 yr return is indeed 3.26% so the effect is most likely caused by your point of using current marketcap weights for calculating historical return for the combo.

BTW, the reason you see different numbers in M* Tax tab vs. the performance tab is that the Tax tab seems to be updated once a month. So you will get the same 3.26 on the performance page if you use the monthly tab instead of the default daily tab which shows 4.08 for VTIAX today. Thanks again for your candid and thorough responses.

User avatar
iceport
Posts: 4151
Joined: Sat Apr 07, 2007 4:29 pm

Re: VTIAX or VTMGX and VEMAX

Post by iceport » Mon Apr 20, 2015 7:17 pm

For some reason, I had to resort to the Wayback Machine to find it, but here's Vanguard's September 24, 2010 announcement that the Total International Stock Index Fund would soon include small caps:
Vanguard plans lower-cost shares, greater diversification for Total International Stock Index Fund
September 24, 2010

Vanguard plans to change the target benchmark of its Total International Stock Index Fund from the MSCI® EAFE + Emerging Markets Index to the MSCI All Country World ex USA Investable Market Index. The fund's new target index covers 98% of the world's non-U.S. markets, including emerging, European, Pacific, and North American regions. The index includes more than 6,000 issues encompassing stocks of large-, mid-, and small-capitalization companies in 44 countries.

"We believe that the new target benchmark offers a better representation of the international equity universe, offering exposure across the capitalization spectrum, including small-cap issues," said Vanguard Chief Investment Officer Gus Sauter, who noted that the benchmark transition will occur during the coming months.
http://web.archive.org/web/201109150930 ... t-09242010

Prior to the late 2010 transition, the fund consisted of a combination of large cap developed and emerging markets indexes.
"Discipline matters more than allocation.” ─William Bernstein

Topic Author
Osp62
Posts: 117
Joined: Sat Oct 04, 2014 9:25 am

Re: VTIAX or VTMGX and VEMAX

Post by Osp62 » Mon Apr 20, 2015 7:25 pm

Thanks Iceport. Greatly appreciate your comments and your taking the time to provide this info.

User avatar
grabiner
Advisory Board
Posts: 25652
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: VTIAX or VTMGX and VEMAX

Post by grabiner » Mon Apr 20, 2015 10:21 pm

Doc wrote:
livesoft wrote:I wanted to make the comment that any numbers reported by M* for "tax-cost" are suspect until proven otherwise. I would not use them myself.
grabiner wrote:Don't use Morningstar's tax data; use the data from the fund provider. Morningstar doesn't always know about qualified dividends, and there are a few other errors.
Both livesoft and grabiner make good points and I don't disagree that M* data can be flawed. It is only useful for ranking very similar funds if at all. However I also question how you could use actual fund's data. While it may be more accurate one would have to go back and look at several years past data to get an idea of how stable the qualified dividends fraction is.
Vanguard has the numbers for every year.

https://personal.vanguard.com/us/insigh ... ncome-2014

Replace 2014 with another year to compare the data for multiple years.

Also, Vanguard reports its own after-tax returns, which are more accurate than Morningstar's because Vanguard does know how much of its own dividends are qualified. For non-Vanguard funds, you may also be able to find after-tax returns from the fund provider.
Another factor that needs to be considered is the "Potential Cap Gains Exposure" which could be important for the future especially if foreign funds are under consideration. This data usually buried in the footnotes in a fund's annual report and can also be found on M*'s tax tab.
Vanguard reports this as "Unrealized Appreciation/Depreciation" on the Distributions tab. However, this is less of a concern for low-turnover index funds, as much of the appreciation will never be realized.
Wiki David Grabiner

User avatar
Doc
Posts: 9307
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: VTIAX or VTMGX and VEMAX

Post by Doc » Tue Apr 21, 2015 9:14 am

grabiner wrote:Vanguard has the numbers for every year.

https://personal.vanguard.com/us/insigh ... ncome-2014

Replace 2014 with another year to compare the data for multiple years.

Also, Vanguard reports its own after-tax returns, which are more accurate than Morningstar's because Vanguard does know how much of its own dividends are qualified. For non-Vanguard funds, you may also be able to find after-tax returns from the fund provider.

(Another factor that needs to be considered is the "Potential Cap Gains Exposure" which could be important for the future especially if foreign funds are under consideration. This data usually buried in the footnotes in a fund's annual report and can also be found on M*'s tax tab.)


Vanguard reports this as "Unrealized Appreciation/Depreciation" on the Distributions tab. However, this is less of a concern for low-turnover index funds, as much of the appreciation will never be realized.
Thanks David, but I thought you were in Maryland not Valley Forge. I didn't know you were Vg's top IT "go to guy". :beer
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

Post Reply