Recreating small cap value world exposure

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
tcrez
Posts: 84
Joined: Mon Mar 01, 2021 11:46 pm

Recreating small cap value world exposure

Post by tcrez »

I want to create VT but for small cap value.

I know about AVUV and AVDV, if these are equivalent to VTI and VEA then I would hold them at weights or 60% and 30%.

I don’t know of an emerging small cap value except AVES, would I hold that one at the 10% that VWO would represent?
DaufuskieNate
Posts: 514
Joined: Wed May 28, 2014 11:53 am

Re: Recreating small cap value world exposure

Post by DaufuskieNate »

DGS is often mentioned in this regard. It is definitely more small cap. While it does have value exposure, it is a dividend fund more than a true value fund.
Topic Author
tcrez
Posts: 84
Joined: Mon Mar 01, 2021 11:46 pm

Re: Recreating small cap value world exposure

Post by tcrez »

It’s a good one when looking at the Morningstar style box.

Must dividend funds be value? Has it ever not been the case?

Also the expense ratio is 0.58%. That seems a bit high right? If the premium is under 1%, you would have very little premium left.
User avatar
whodidntante
Posts: 11013
Joined: Thu Jan 21, 2016 11:11 pm
Location: outside the echo chamber

Re: Recreating small cap value world exposure

Post by whodidntante »

tcrez wrote: Fri Jul 29, 2022 6:43 pm It’s a good one when looking at the Morningstar style box.

Must dividend funds be value? Has it ever not been the case?

Also the expense ratio is 0.58%. That seems a bit high right? If the premium is under 1%, you would have very little premium left.
I do not recommend owning DGS for EM SCV because it's too expensive, and as my astute colleague above noted, not a value fund. High dividend is not equivalent to value. It's true that high dividend is correlated to value. But that's because a lot of value companies can't think of anything better to do with their capital than pay it out to shareholders. Still, it's a poor proxy. Avoid it. You'll end up owning some real crap that is not supported by the best models that we have.

I personally don't find EM SCV worth targetting. It's a tiny slice of already tiny markets. And the trading costs for EM SCV equities border on horrendous. You noted the high ER, and that's valid. But you also have to consider if there is anything left for you after all costs. The trading costs, foreign tax cost, and low percentage of qualified dividends are worth consideration. My advice is to let that premium live in the models. Momentum looks great in models also, but good luck capturing it.

The droid you are looking for does not exist. As reasonable alternatives, I would use DFEV or IEMG.
Nathan Drake
Posts: 3650
Joined: Mon Apr 11, 2011 12:28 am

Re: Recreating small cap value world exposure

Post by Nathan Drake »

AVUV - 50%
AVDV - 35%
AVES - 15%

something like that, doesn't need to be exact. You are getting a global small cap value exposure with this allocation.

You can tweak the percentages to taste.
20% VOO | 20% VXUS | 20% AVUV | 20% AVDV | 20% AVES
comeinvest
Posts: 1260
Joined: Mon Mar 12, 2012 6:57 pm

Re: Recreating small cap value world exposure

Post by comeinvest »

whodidntante wrote: Fri Jul 29, 2022 7:04 pm
tcrez wrote: Fri Jul 29, 2022 6:43 pm It’s a good one when looking at the Morningstar style box.

Must dividend funds be value? Has it ever not been the case?

Also the expense ratio is 0.58%. That seems a bit high right? If the premium is under 1%, you would have very little premium left.
I do not recommend owning DGS for EM SCV because it's too expensive, and as my astute colleague above noted, not a value fund. High dividend is not equivalent to value. It's true that high dividend is correlated to value. But that's because a lot of value companies can't think of anything better to do with their capital than pay it out to shareholders. Still, it's a poor proxy. Avoid it. You'll end up owning some real crap that is not supported by the best models that we have.

I personally don't find EM SCV worth targetting. It's a tiny slice of already tiny markets. And the trading costs for EM SCV equities border on horrendous. You noted the high ER, and that's valid. But you also have to consider if there is anything left for you after all costs. The trading costs, foreign tax cost, and low percentage of qualified dividends are worth consideration. My advice is to let that premium live in the models. Momentum looks great in models also, but good luck capturing it.

The droid you are looking for does not exist. As reasonable alternatives, I would use DFEV or IEMG.
I don't disagree with your assessment and the all-in cost is indeed horrendous, but a counterargument is that the SCV premium according to some models is supposed to be higher for small caps, and higher for foreign companies especially EM. Also, last time I checked, DGS applies some intriguing "quality factor" and other filters, on top of the heavy value tilt which they implement via a dividend screen, which seems to be one possible way and a somewhat efficient way to implement it due to the high correlation of dividends with value, whether or not it is the most orthodox way according to some academic theories. I personally am still deliberating between DGS, AVES, or no EM SCV tilt at all.
DGS outperformed both AVES and IEMG since AVES' inception 1.5 years ago, but of course that proves little to nothing because of the short time span.
For what it's worth, DGS outperformed VWO (emerging market index fund) by more than 2% p.a. since inception in 2008, with less max drawdown than VWO. That's kind of confidence instilling, isn't it?
https://www.portfoliovisualizer.com/bac ... ion2_2=100

There is also a point of view that EM small caps are more independent and uncorrelated with DM than EM large caps, and on top that they are less government controlled than EM LC. Possibly a better diversifier than EM LC.
User avatar
burritoLover
Posts: 2543
Joined: Sun Jul 05, 2020 12:13 pm

Re: Recreating small cap value world exposure

Post by burritoLover »

comeinvest wrote: Thu Aug 04, 2022 3:03 am
whodidntante wrote: Fri Jul 29, 2022 7:04 pm
tcrez wrote: Fri Jul 29, 2022 6:43 pm It’s a good one when looking at the Morningstar style box.

Must dividend funds be value? Has it ever not been the case?

Also the expense ratio is 0.58%. That seems a bit high right? If the premium is under 1%, you would have very little premium left.
I do not recommend owning DGS for EM SCV because it's too expensive, and as my astute colleague above noted, not a value fund. High dividend is not equivalent to value. It's true that high dividend is correlated to value. But that's because a lot of value companies can't think of anything better to do with their capital than pay it out to shareholders. Still, it's a poor proxy. Avoid it. You'll end up owning some real crap that is not supported by the best models that we have.

I personally don't find EM SCV worth targetting. It's a tiny slice of already tiny markets. And the trading costs for EM SCV equities border on horrendous. You noted the high ER, and that's valid. But you also have to consider if there is anything left for you after all costs. The trading costs, foreign tax cost, and low percentage of qualified dividends are worth consideration. My advice is to let that premium live in the models. Momentum looks great in models also, but good luck capturing it.

The droid you are looking for does not exist. As reasonable alternatives, I would use DFEV or IEMG.
I don't disagree with your assessment and the all-in cost is indeed horrendous, but a counterargument is that the SCV premium according to some models is supposed to be higher for small caps, and higher for foreign companies especially EM. Also, last time I checked, DGS applies some intriguing "quality factor" and other filters, on top of the heavy value tilt which they implement via a dividend screen, which seems to be one possible way and a somewhat efficient way to implement it due to the high correlation of dividends with value, whether or not it is the most orthodox way according to some academic theories. I personally am still deliberating between DGS, AVES, or no EM SCV tilt at all.
DGS outperformed both AVES and IEMG since AVES' inception 1.5 years ago, but of course that proves little to nothing because of the short time span.
For what it's worth, DGS outperformed VWO (emerging market index fund) by more than 2% p.a. since inception in 2008, with less max drawdown than VWO. That's kind of confidence instilling, isn't it?
https://www.portfoliovisualizer.com/bac ... ion2_2=100

There is also a point of view that EM small caps are more independent and uncorrelated with DM than EM large caps, and on top that they are less government controlled than EM LC. Possibly a better diversifier than EM LC.
Factor models don't dictate what area of the world will have a greater value premium. EM value happened to have outperformed ex-US developed and US value for a period of time (maybe several decades). That is meaningless going forward as is the backtest of DSG vs EM market.
comeinvest
Posts: 1260
Joined: Mon Mar 12, 2012 6:57 pm

Re: Recreating small cap value world exposure

Post by comeinvest »

burritoLover wrote: Thu Aug 04, 2022 6:26 am
comeinvest wrote: Thu Aug 04, 2022 3:03 am
whodidntante wrote: Fri Jul 29, 2022 7:04 pm
tcrez wrote: Fri Jul 29, 2022 6:43 pm It’s a good one when looking at the Morningstar style box.

Must dividend funds be value? Has it ever not been the case?

Also the expense ratio is 0.58%. That seems a bit high right? If the premium is under 1%, you would have very little premium left.
I do not recommend owning DGS for EM SCV because it's too expensive, and as my astute colleague above noted, not a value fund. High dividend is not equivalent to value. It's true that high dividend is correlated to value. But that's because a lot of value companies can't think of anything better to do with their capital than pay it out to shareholders. Still, it's a poor proxy. Avoid it. You'll end up owning some real crap that is not supported by the best models that we have.

I personally don't find EM SCV worth targetting. It's a tiny slice of already tiny markets. And the trading costs for EM SCV equities border on horrendous. You noted the high ER, and that's valid. But you also have to consider if there is anything left for you after all costs. The trading costs, foreign tax cost, and low percentage of qualified dividends are worth consideration. My advice is to let that premium live in the models. Momentum looks great in models also, but good luck capturing it.

The droid you are looking for does not exist. As reasonable alternatives, I would use DFEV or IEMG.
I don't disagree with your assessment and the all-in cost is indeed horrendous, but a counterargument is that the SCV premium according to some models is supposed to be higher for small caps, and higher for foreign companies especially EM. Also, last time I checked, DGS applies some intriguing "quality factor" and other filters, on top of the heavy value tilt which they implement via a dividend screen, which seems to be one possible way and a somewhat efficient way to implement it due to the high correlation of dividends with value, whether or not it is the most orthodox way according to some academic theories. I personally am still deliberating between DGS, AVES, or no EM SCV tilt at all.
DGS outperformed both AVES and IEMG since AVES' inception 1.5 years ago, but of course that proves little to nothing because of the short time span.
For what it's worth, DGS outperformed VWO (emerging market index fund) by more than 2% p.a. since inception in 2008, with less max drawdown than VWO. That's kind of confidence instilling, isn't it?
https://www.portfoliovisualizer.com/bac ... ion2_2=100

There is also a point of view that EM small caps are more independent and uncorrelated with DM than EM large caps, and on top that they are less government controlled than EM LC. Possibly a better diversifier than EM LC.
Factor models don't dictate what area of the world will have a greater value premium. EM value happened to have outperformed ex-US developed and US value for a period of time (maybe several decades). That is meaningless going forward as is the backtest of DSG vs EM market.
No backtest says anything about the future per se, nobody in the world knows if any current factor theories will be relevant in the future, and I agree that an anecdotal 14 years backtest does not validate any theory. In short, nobody knows nothing for sure about the future. Everything that is ever said about the future, are nothing but educated guesses. Nevertheless, we have to make choices now, and those choices will affect our future investment returns. One of those choices is to allocate some money to EM SCV or to go entirely with the EM large cap index. I do think that a 2% p.a. outperformance over 14 years, along with the theoretical framework that scientist created with much longer data sets worldwide, and along with diversification arguments, instills some confidence that a portfolio that diverts some EM index allocation to DGS might have and edge in terms of risk-adjusted returns.
Just like whodidntante, I am concerned about the all-in cost involved with DGS.
User avatar
burritoLover
Posts: 2543
Joined: Sun Jul 05, 2020 12:13 pm

Re: Recreating small cap value world exposure

Post by burritoLover »

comeinvest wrote: Thu Aug 04, 2022 8:47 pm
burritoLover wrote: Thu Aug 04, 2022 6:26 am
comeinvest wrote: Thu Aug 04, 2022 3:03 am
whodidntante wrote: Fri Jul 29, 2022 7:04 pm
tcrez wrote: Fri Jul 29, 2022 6:43 pm It’s a good one when looking at the Morningstar style box.

Must dividend funds be value? Has it ever not been the case?

Also the expense ratio is 0.58%. That seems a bit high right? If the premium is under 1%, you would have very little premium left.
I do not recommend owning DGS for EM SCV because it's too expensive, and as my astute colleague above noted, not a value fund. High dividend is not equivalent to value. It's true that high dividend is correlated to value. But that's because a lot of value companies can't think of anything better to do with their capital than pay it out to shareholders. Still, it's a poor proxy. Avoid it. You'll end up owning some real crap that is not supported by the best models that we have.

I personally don't find EM SCV worth targetting. It's a tiny slice of already tiny markets. And the trading costs for EM SCV equities border on horrendous. You noted the high ER, and that's valid. But you also have to consider if there is anything left for you after all costs. The trading costs, foreign tax cost, and low percentage of qualified dividends are worth consideration. My advice is to let that premium live in the models. Momentum looks great in models also, but good luck capturing it.

The droid you are looking for does not exist. As reasonable alternatives, I would use DFEV or IEMG.
I don't disagree with your assessment and the all-in cost is indeed horrendous, but a counterargument is that the SCV premium according to some models is supposed to be higher for small caps, and higher for foreign companies especially EM. Also, last time I checked, DGS applies some intriguing "quality factor" and other filters, on top of the heavy value tilt which they implement via a dividend screen, which seems to be one possible way and a somewhat efficient way to implement it due to the high correlation of dividends with value, whether or not it is the most orthodox way according to some academic theories. I personally am still deliberating between DGS, AVES, or no EM SCV tilt at all.
DGS outperformed both AVES and IEMG since AVES' inception 1.5 years ago, but of course that proves little to nothing because of the short time span.
For what it's worth, DGS outperformed VWO (emerging market index fund) by more than 2% p.a. since inception in 2008, with less max drawdown than VWO. That's kind of confidence instilling, isn't it?
https://www.portfoliovisualizer.com/bac ... ion2_2=100

There is also a point of view that EM small caps are more independent and uncorrelated with DM than EM large caps, and on top that they are less government controlled than EM LC. Possibly a better diversifier than EM LC.
Factor models don't dictate what area of the world will have a greater value premium. EM value happened to have outperformed ex-US developed and US value for a period of time (maybe several decades). That is meaningless going forward as is the backtest of DSG vs EM market.
No backtest says anything about the future per se, nobody in the world knows if any current factor theories will be relevant in the future, and I agree that an anecdotal 14 years backtest does not validate any theory. In short, nobody knows nothing for sure about the future. Everything that is ever said about the future, are nothing but educated guesses. Nevertheless, we have to make choices now, and those choices will affect our future investment returns. One of those choices is to allocate some money to EM SCV or to go entirely with the EM large cap index. I do think that a 2% p.a. outperformance over 14 years, along with the theoretical framework that scientist created with much longer data sets worldwide, and along with diversification arguments, instills some confidence that a portfolio that diverts some EM index allocation to DGS might have and edge in terms of risk-adjusted returns.
Just like whodidntante, I am concerned about the all-in cost involved with DGS.
I'm not sure why you wouldn't just go with AVES if you want to target the value factor - an actual EM value fund - not a dividend-only fund like DGS. The loadings on size are about the same as well. And the ER is cheaper (0.36%). Yes, we don't know if the premium for small and value will even exist going forward (or be greatly reduced) but if your purpose is to invest in factors, then I would do that directly and Avantis appears to have a sound methodology.
comeinvest
Posts: 1260
Joined: Mon Mar 12, 2012 6:57 pm

Re: Recreating small cap value world exposure

Post by comeinvest »

burritoLover wrote: Fri Aug 05, 2022 6:12 am
comeinvest wrote: Thu Aug 04, 2022 8:47 pm
burritoLover wrote: Thu Aug 04, 2022 6:26 am
comeinvest wrote: Thu Aug 04, 2022 3:03 am
whodidntante wrote: Fri Jul 29, 2022 7:04 pm

I do not recommend owning DGS for EM SCV because it's too expensive, and as my astute colleague above noted, not a value fund. High dividend is not equivalent to value. It's true that high dividend is correlated to value. But that's because a lot of value companies can't think of anything better to do with their capital than pay it out to shareholders. Still, it's a poor proxy. Avoid it. You'll end up owning some real crap that is not supported by the best models that we have.

I personally don't find EM SCV worth targetting. It's a tiny slice of already tiny markets. And the trading costs for EM SCV equities border on horrendous. You noted the high ER, and that's valid. But you also have to consider if there is anything left for you after all costs. The trading costs, foreign tax cost, and low percentage of qualified dividends are worth consideration. My advice is to let that premium live in the models. Momentum looks great in models also, but good luck capturing it.

The droid you are looking for does not exist. As reasonable alternatives, I would use DFEV or IEMG.
I don't disagree with your assessment and the all-in cost is indeed horrendous, but a counterargument is that the SCV premium according to some models is supposed to be higher for small caps, and higher for foreign companies especially EM. Also, last time I checked, DGS applies some intriguing "quality factor" and other filters, on top of the heavy value tilt which they implement via a dividend screen, which seems to be one possible way and a somewhat efficient way to implement it due to the high correlation of dividends with value, whether or not it is the most orthodox way according to some academic theories. I personally am still deliberating between DGS, AVES, or no EM SCV tilt at all.
DGS outperformed both AVES and IEMG since AVES' inception 1.5 years ago, but of course that proves little to nothing because of the short time span.
For what it's worth, DGS outperformed VWO (emerging market index fund) by more than 2% p.a. since inception in 2008, with less max drawdown than VWO. That's kind of confidence instilling, isn't it?
https://www.portfoliovisualizer.com/bac ... ion2_2=100

There is also a point of view that EM small caps are more independent and uncorrelated with DM than EM large caps, and on top that they are less government controlled than EM LC. Possibly a better diversifier than EM LC.
Factor models don't dictate what area of the world will have a greater value premium. EM value happened to have outperformed ex-US developed and US value for a period of time (maybe several decades). That is meaningless going forward as is the backtest of DSG vs EM market.
No backtest says anything about the future per se, nobody in the world knows if any current factor theories will be relevant in the future, and I agree that an anecdotal 14 years backtest does not validate any theory. In short, nobody knows nothing for sure about the future. Everything that is ever said about the future, are nothing but educated guesses. Nevertheless, we have to make choices now, and those choices will affect our future investment returns. One of those choices is to allocate some money to EM SCV or to go entirely with the EM large cap index. I do think that a 2% p.a. outperformance over 14 years, along with the theoretical framework that scientist created with much longer data sets worldwide, and along with diversification arguments, instills some confidence that a portfolio that diverts some EM index allocation to DGS might have and edge in terms of risk-adjusted returns.
Just like whodidntante, I am concerned about the all-in cost involved with DGS.
I'm not sure why you wouldn't just go with AVES if you want to target the value factor - an actual EM value fund - not a dividend-only fund like DGS. The loadings on size are about the same as well. And the ER is cheaper (0.36%). Yes, we don't know if the premium for small and value will even exist going forward (or be greatly reduced) but if your purpose is to invest in factors, then I would do that directly and Avantis appears to have a sound methodology.
That's exactly what I did. I directed new money to AVES, and converted some of my DGS to AVES. I don't want to do a 180 degree turn right now, as DGS has some potential advantages, like smaller average market cap of constituents. They also have some quality filters, and I think the dividend strategy is not as bad as some factor purists try to make us believe, just because it doesn't fit their orthodox theories. Although there are some companies that pay dividends with no earnings, and others that pay nothing with plenty of earnings, "value" and "dividends" result in very similar portfolios. I agree however that AVES is probably better and I will probably keep migrating assets from DGS to AVES, maybe keeping a smaller allocation to DGS for diversification of strategies. AVES has a portfolio dividend yield of 5.5%, DGS 8.5%, which is a bit insane and will naturally result in a higher drag from withholding taxes. On the other hand, it was argued that companies that pay cash are less likely to have a management "agency" problem, a widely held concern especially in EM. I think I would go with the 5.5% though.
Last edited by comeinvest on Fri Aug 05, 2022 6:24 pm, edited 1 time in total.
User avatar
burritoLover
Posts: 2543
Joined: Sun Jul 05, 2020 12:13 pm

Re: Recreating small cap value world exposure

Post by burritoLover »

comeinvest wrote: Fri Aug 05, 2022 6:14 pm
burritoLover wrote: Fri Aug 05, 2022 6:12 am
comeinvest wrote: Thu Aug 04, 2022 8:47 pm
burritoLover wrote: Thu Aug 04, 2022 6:26 am
comeinvest wrote: Thu Aug 04, 2022 3:03 am

I don't disagree with your assessment and the all-in cost is indeed horrendous, but a counterargument is that the SCV premium according to some models is supposed to be higher for small caps, and higher for foreign companies especially EM. Also, last time I checked, DGS applies some intriguing "quality factor" and other filters, on top of the heavy value tilt which they implement via a dividend screen, which seems to be one possible way and a somewhat efficient way to implement it due to the high correlation of dividends with value, whether or not it is the most orthodox way according to some academic theories. I personally am still deliberating between DGS, AVES, or no EM SCV tilt at all.
DGS outperformed both AVES and IEMG since AVES' inception 1.5 years ago, but of course that proves little to nothing because of the short time span.
For what it's worth, DGS outperformed VWO (emerging market index fund) by more than 2% p.a. since inception in 2008, with less max drawdown than VWO. That's kind of confidence instilling, isn't it?
https://www.portfoliovisualizer.com/bac ... ion2_2=100

There is also a point of view that EM small caps are more independent and uncorrelated with DM than EM large caps, and on top that they are less government controlled than EM LC. Possibly a better diversifier than EM LC.
Factor models don't dictate what area of the world will have a greater value premium. EM value happened to have outperformed ex-US developed and US value for a period of time (maybe several decades). That is meaningless going forward as is the backtest of DSG vs EM market.
No backtest says anything about the future per se, nobody in the world knows if any current factor theories will be relevant in the future, and I agree that an anecdotal 14 years backtest does not validate any theory. In short, nobody knows nothing for sure about the future. Everything that is ever said about the future, are nothing but educated guesses. Nevertheless, we have to make choices now, and those choices will affect our future investment returns. One of those choices is to allocate some money to EM SCV or to go entirely with the EM large cap index. I do think that a 2% p.a. outperformance over 14 years, along with the theoretical framework that scientist created with much longer data sets worldwide, and along with diversification arguments, instills some confidence that a portfolio that diverts some EM index allocation to DGS might have and edge in terms of risk-adjusted returns.
Just like whodidntante, I am concerned about the all-in cost involved with DGS.
I'm not sure why you wouldn't just go with AVES if you want to target the value factor - an actual EM value fund - not a dividend-only fund like DGS. The loadings on size are about the same as well. And the ER is cheaper (0.36%). Yes, we don't know if the premium for small and value will even exist going forward (or be greatly reduced) but if your purpose is to invest in factors, then I would do that directly and Avantis appears to have a sound methodology.
That's exactly what I did. I directed new money to AVES, and converted some of my DGS to AVES. I don't want to do a 180 degree turn right now, as DGS has some potential advantages, like smaller average market cap of constituents. They also have some quality filters, and I think the dividend strategy is not as bad as some factor purists try to make us believe, just because it doesn't fit their orthodox theories. Although there are some companies that pay dividends with no earnings, and others that pay nothing with plenty of earnings, "value" and "dividends" result in very similar portfolios. I do agree however that AVES is probably better and I will probably keep migrating assets from DGS to AVES, maybe keeping a smaller allocation to DGS for diversification of strategies.
Ah I see. Well, if you feel strongly about DGS, you could always just shoot for a 50/50 DGS/AVES allocation.
Post Reply