Do DFA Funds have lower trading costs than Index Funds?
-
- Posts: 21
- Joined: Mon Oct 09, 2017 12:09 am
Do DFA Funds have lower trading costs than Index Funds?
I recently attended a DFA foundations conference in which they described their factor based investment strategy
One thing mentioned at the conference which really stood out to me was that they stated the overall trading costs of their funds was lower than both active and index fund counterparts. In regards to active they stated DFA funds have lower turnover rates, which leads to lower overall cost per definition. In regards to index funds they stated that DFA funds had much lower Implicit costs, which they defined as costs related to market moving and costs due to the bid ask spread.
Their explanation for this was as follows:
When trading on the market one can see it like a negotiation between sellers and buyers, in these negotiations there are three main factors that each party puts different levels of priority on. Time, which means how directly you are in need of the asset, Quantity, which they defined as how specific you are regarding what you want, for example do you want one specific small cap stock or are you indifferent between buying 5 or 10 different stocks. And finally price, which they defined as the price you in the end are willing to settle for. They stated you can prioritize two of these factors at a time, meaning you have to be flexible in the third factor and "accept what you get" there.
They stated that index funds in general rebalance once every year when the index they track rebalances. When this occurs the index funds need to shift their allocations accordingly and quickly to minimize tracking error. This leads to non flexibility in regards to time (since it needs to occur quickly) and non flexibility in regards to quantity (since they need to buy exactly what is included in the underlying index they are tracking). Due to this they need to be flexible in regards to the third factor namely price. Meaning that even if many index funds rebalance during the same time period and move the market, they are all still forced to buy, in doing so they need to cross the bid ask spread making them lose out in that regard aswell.
They then stated that since they gave their traders a high degree of flexibility in regards to quantity ie. which specific stock to buy (as long as the stock fulfilled the factor criterias, which alot of stocks tended to do, it was a buy) as well as flexibility in regards to time (since they do not track a specific index and therefore do not need to shift allocations quickly to minimize tracking error). DFA can be less flexible in regards to price, meaning they get more beneficial prices and therefore are able to have lower implicit trading costs. This is possible since they dont ever need to cross the bid ask spread and can instead wait for the price to come to them ie. place an order on the bid and wait for the transaction to go through. They can also buy in smaller quantities over a long time period meaning they do not move the market.
They then showed a study they had conducted which highlighted how they in general buy for a lower price and sell for a higher price than other funds. They also showed how they on average buy at a lower price than the two adjacent trades, and sell for a higher price than the two adjacent trades.
It was very interesting to listen to and I was hoping to get some Bogelheads thoughts on this
All input is greatly appreciated!
One thing mentioned at the conference which really stood out to me was that they stated the overall trading costs of their funds was lower than both active and index fund counterparts. In regards to active they stated DFA funds have lower turnover rates, which leads to lower overall cost per definition. In regards to index funds they stated that DFA funds had much lower Implicit costs, which they defined as costs related to market moving and costs due to the bid ask spread.
Their explanation for this was as follows:
When trading on the market one can see it like a negotiation between sellers and buyers, in these negotiations there are three main factors that each party puts different levels of priority on. Time, which means how directly you are in need of the asset, Quantity, which they defined as how specific you are regarding what you want, for example do you want one specific small cap stock or are you indifferent between buying 5 or 10 different stocks. And finally price, which they defined as the price you in the end are willing to settle for. They stated you can prioritize two of these factors at a time, meaning you have to be flexible in the third factor and "accept what you get" there.
They stated that index funds in general rebalance once every year when the index they track rebalances. When this occurs the index funds need to shift their allocations accordingly and quickly to minimize tracking error. This leads to non flexibility in regards to time (since it needs to occur quickly) and non flexibility in regards to quantity (since they need to buy exactly what is included in the underlying index they are tracking). Due to this they need to be flexible in regards to the third factor namely price. Meaning that even if many index funds rebalance during the same time period and move the market, they are all still forced to buy, in doing so they need to cross the bid ask spread making them lose out in that regard aswell.
They then stated that since they gave their traders a high degree of flexibility in regards to quantity ie. which specific stock to buy (as long as the stock fulfilled the factor criterias, which alot of stocks tended to do, it was a buy) as well as flexibility in regards to time (since they do not track a specific index and therefore do not need to shift allocations quickly to minimize tracking error). DFA can be less flexible in regards to price, meaning they get more beneficial prices and therefore are able to have lower implicit trading costs. This is possible since they dont ever need to cross the bid ask spread and can instead wait for the price to come to them ie. place an order on the bid and wait for the transaction to go through. They can also buy in smaller quantities over a long time period meaning they do not move the market.
They then showed a study they had conducted which highlighted how they in general buy for a lower price and sell for a higher price than other funds. They also showed how they on average buy at a lower price than the two adjacent trades, and sell for a higher price than the two adjacent trades.
It was very interesting to listen to and I was hoping to get some Bogelheads thoughts on this
All input is greatly appreciated!
Re: Do DFA Funds have lower trading costs than Index Funds?
Following, my DFA advisor says the sameFlugugrubah wrote: ↑Sun Dec 10, 2017 1:51 pm I recently attended a DFA foundations conference in which they described their factor based investment strategy
One thing mentioned at the conference which really stood out to me was that they stated the overall trading costs of their funds was lower than both active and index fund counterparts. In regards to active they stated DFA funds have lower turnover rates, which leads to lower overall cost per definition. In regards to index funds they stated that DFA funds had much lower Implicit costs, which they defined as costs related to market moving and costs due to the bid ask spread.
Their explanation for this was as follows:
When trading on the market one can see it like a negotiation between sellers and buyers, in these negotiations there are three main factors that each party puts different levels of priority on. Time, which means how directly you are in need of the asset, Quantity, which they defined as how specific you are regarding what you want, for example do you want one specific small cap stock or are you indifferent between buying 5 or 10 different stocks. And finally price, which they defined as the price you in the end are willing to settle for. They stated you can prioritize two of these factors at a time, meaning you have to be flexible in the third factor and "accept what you get" there.
They stated that index funds in general rebalance once every year when the index they track rebalances. When this occurs the index funds need to shift their allocations accordingly and quickly to minimize tracking error. This leads to non flexibility in regards to time (since it needs to occur quickly) and non flexibility in regards to quantity (since they need to buy exactly what is included in the underlying index they are tracking). Due to this they need to be flexible in regards to the third factor namely price. Meaning that even if many index funds rebalance during the same time period and move the market, they are all still forced to buy, in doing so they need to cross the bid ask spread making them lose out in that regard aswell.
They then stated that since they gave their traders a high degree of flexibility in regards to quantity ie. which specific stock to buy (as long as the stock fulfilled the factor criterias, which alot of stocks tended to do, it was a buy) as well as flexibility in regards to time (since they do not track a specific index and therefore do not need to shift allocations quickly to minimize tracking error). DFA can be less flexible in regards to price, meaning they get more beneficial prices and therefore are able to have lower implicit trading costs. This is possible since they dont ever need to cross the bid ask spread and can instead wait for the price to come to them ie. place an order on the bid and wait for the transaction to go through. They can also buy in smaller quantities over a long time period meaning they do not move the market.
They then showed a study they had conducted which highlighted how they in general buy for a lower price and sell for a higher price than other funds. They also showed how they on average buy at a lower price than the two adjacent trades, and sell for a higher price than the two adjacent trades.
It was very interesting to listen to and I was hoping to get some Bogelheads thoughts on this
All input is greatly appreciated!
Re: Do DFA Funds have lower trading costs than Index Funds?
I'm sure they do better when comparing to a universe of index funds. However, that universe contains the good and the bad index funds. I really don't care too much about the universe of index funds because I only use a few index funds sponsored by Vanguard. So I don't think if they did a comparison to Vanguard alone that they would be much different. What do you think?
Re: Do DFA Funds have lower trading costs than Index Funds?
Sure, this has been one of the points DFA has made for a long time.
This is one of the advantages to not following an index, really. Decent index fund managers are cognizant of trading costs too, but they have a mandate on performance (keeping transaction costs low) as well as tracking consistency with respect to the index, which they thus have to trade around in an environment where other traders know what they're up to. If you care less about maintaining your ideal portfolio* and more about trading frictions, then you can reduce trading frictions.
*for DFA's case, this may mean slightly lower factor exposures, not having exactly the set of stocks that are being targeted, though as you mention here they downplay this and say that many stocks are available that fit the criteria, which is largely probably true
DFA is not the only one doing this, but it should be an advantage over index funds (at the expense of less transparency, not having a public index to benchmark the fund against, etc.) and certainly over many active funds. If a discretionary active manager wants to buy stock X or really wants to sell stock Y, they may want to move more aggressively at that moment, at the expense of effectively eating higher transaction costs. The active manager would say that their judgment of the timing is more important than that additional cost for more timely execution, but if you generally don't believe in their judgement of the timing, then from that perspective they're just eating higher costs.
So yeah, DFA can have lower implicit costs from trading. They just conveniently have higher explicit costs than many index funds, offsetting those implicit costs and more, potentially. I guess it would depend on the exact trading costs for the given market targeted, ERs, and so on. Also, in some categories DFA could have higher turnover to deal with. Even if more efficient trading per dollar, the net effect could be larger if there are more dollars being turned over.
This is one of the advantages to not following an index, really. Decent index fund managers are cognizant of trading costs too, but they have a mandate on performance (keeping transaction costs low) as well as tracking consistency with respect to the index, which they thus have to trade around in an environment where other traders know what they're up to. If you care less about maintaining your ideal portfolio* and more about trading frictions, then you can reduce trading frictions.
*for DFA's case, this may mean slightly lower factor exposures, not having exactly the set of stocks that are being targeted, though as you mention here they downplay this and say that many stocks are available that fit the criteria, which is largely probably true
DFA is not the only one doing this, but it should be an advantage over index funds (at the expense of less transparency, not having a public index to benchmark the fund against, etc.) and certainly over many active funds. If a discretionary active manager wants to buy stock X or really wants to sell stock Y, they may want to move more aggressively at that moment, at the expense of effectively eating higher transaction costs. The active manager would say that their judgment of the timing is more important than that additional cost for more timely execution, but if you generally don't believe in their judgement of the timing, then from that perspective they're just eating higher costs.
So yeah, DFA can have lower implicit costs from trading. They just conveniently have higher explicit costs than many index funds, offsetting those implicit costs and more, potentially. I guess it would depend on the exact trading costs for the given market targeted, ERs, and so on. Also, in some categories DFA could have higher turnover to deal with. Even if more efficient trading per dollar, the net effect could be larger if there are more dollars being turned over.
Last edited by lack_ey on Sun Dec 10, 2017 2:07 pm, edited 1 time in total.
- triceratop
- Posts: 5838
- Joined: Tue Aug 04, 2015 8:20 pm
- Location: la la land
Re: Do DFA Funds have lower trading costs than Index Funds?
This may be true, but did DFA mention what index funds they are comparing to? Was it total market funds, or small cap value funds? Rebalancing effects may be more important for some indexing approaches than others. Did DFA say these costs show up in fund tracking error, or does the index itself hide the cost? I don't know the answers to these questions, I'm genuinely curious.
The reason I ask that last question: here are the last 10-year returns as reported in the annual report for Vanguard's Total Market fund:
Always useful to understand that while DFA's critique might hold for some index investors, does it impact you, the (possibly) 3-fund investor?
The reason I ask that last question: here are the last 10-year returns as reported in the annual report for Vanguard's Total Market fund:
Code: Select all
-- 1yr 5yr 10yr
TSM Fund Admiral Shares 12.66% 14.62% 7.23%
Spliced TSM Index 12.68 14.64 7.23
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."
Re: Do DFA Funds have lower trading costs than Index Funds?
As a small-cap and value-tilted shop, I would expect DFA to kick performance butt with their DFSVX small-cap value fund. Sure, it is not exactly the asset allocation of something like the Vanguard small-cap value fund, but [year-to-date] 5.8% for DFA versus 10.9% for Vanguard means that whatever DFA is doing to patiently trade is not going to make me like them. So given this shortfall, the DFA marketing folks are out in force telling people to manage their expectations and other good things about DFA. I don't fall for it myself.
Last edited by livesoft on Sun Dec 10, 2017 2:30 pm, edited 1 time in total.
- whodidntante
- Posts: 13090
- Joined: Thu Jan 21, 2016 10:11 pm
- Location: outside the echo chamber
Re: Do DFA Funds have lower trading costs than Index Funds?
Yes, DFA's approach does help when trading small, illiquid securities. There are a lot of them and taken together they comprise about 5% of the market, and a lot of them don't really trade all that well. So a mandate to track an index isn't helpful. Small, micro, and some EM stocks will benefit from this. It matters less when you get to large float stocks.
- nisiprius
- Advisory Board
- Posts: 52107
- Joined: Thu Jul 26, 2007 9:33 am
- Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry
Re: Do DFA Funds have lower trading costs than Index Funds?
Heaven knows I'm a DFA skeptic--not that I think they're wrong, exactly, but I think they are trying to make a small difference look big and talk about secret miracle ingredients... but where are you getting those 5.8% versus 10.9% numbers?livesoft wrote: ↑Sun Dec 10, 2017 2:05 pm As a small-cap and value-tilted shop, I would expect DFA to kick performance butt with their DFSVX small-cap value fund. Sure, it is not exactly the asset allocation of something like the Vanguard small-cap value fund, but 5.8% for DFA versus 10.9% for Vanguard means that whatever DFA is doing to patiently trade is not going to make me like them. So given this shortfall, the DFA marketing folks are out in force telling people to manage their expectations and other good things about DFA. I don't fall for it myself.
Source
Since inception on 5/21/1998, I calculate 9.75% for DFSVX, DFA U.S. Small-Cap Value Portfolio and 8.93% for VISVX, Vanguard Small-Cap Value Index. That falls just beautifully in the grey area of endless argument; PortfolioVisualizer is showing a slightly standard deviation for the DFA fund, and thus the Sharpe ratios are 0.48 for DFA, 0.46 Vanguard... which depending on your point of view shows "DFA was better" or "they were about the same."
Transaction costs should be lower with patient trading, that's the point... but I don't know if there's any way to come with actual numbers, since the regulations that require disclosure of various things don't requires disclosure of that thing. And there is at least a suspicion that Vanguard's size gives them the clout to get favorable trading arrangements, and that Vanguard has some transactional skill of its own, as shown by the fact that in many cases the fund performance lags the index by noticeably less than the expense ratio.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Re: Do DFA Funds have lower trading costs than Index Funds?
From morningstar.com. The numbers I reported are YTD and not from some ancient past. Your chart makes me want to ask "Dear DFA, what have you done for me lately?"
-
- Posts: 21
- Joined: Mon Oct 09, 2017 12:09 am
Re: Do DFA Funds have lower trading costs than Index Funds?
I am not a DFA investor myself however I do believe in alot of what their strategy stands for, as they put a focus on finding robust premiums that deliver value after they have been implemented.nisiprius wrote: ↑Sun Dec 10, 2017 2:25 pmHeaven knows I'm a DFA skeptic--not that I think they're wrong, exactly, but I think they are trying to make a small difference look big and talk about secret miracle ingredients... but where are you getting those 5.8% versus 10.9% numbers?livesoft wrote: ↑Sun Dec 10, 2017 2:05 pm As a small-cap and value-tilted shop, I would expect DFA to kick performance butt with their DFSVX small-cap value fund. Sure, it is not exactly the asset allocation of something like the Vanguard small-cap value fund, but 5.8% for DFA versus 10.9% for Vanguard means that whatever DFA is doing to patiently trade is not going to make me like them. So given this shortfall, the DFA marketing folks are out in force telling people to manage their expectations and other good things about DFA. I don't fall for it myself.
Source
Since inception on 5/21/1998, I calculate 9.75% for DFSVX, DFA U.S. Small-Cap Value Portfolio and 8.93% for VISVX, Vanguard Small-Cap Value Index. That falls just beautifully in the grey area of endless argument; PortfolioVisualizer is showing a slightly standard deviation for the DFA fund, and thus the Sharpe ratios are 0.48 for DFA, 0.46 Vanguard... which depending on your point of view shows "DFA was better" or "they were about the same."
Transaction costs should be lower with patient trading, that's the point... but I don't know if there's any way to come with actual numbers, since the regulations that require disclosure of various things don't requires disclosure of that thing. And there is at least a suspicion that Vanguard's size gives them the clout to get favorable trading arrangements, and that Vanguard has some transactional skill of its own, as shown by the fact that in many cases the fund performance lags the index by noticeably less than the expense ratio.
I am aware that there are many questions regarding if DFAs factor investing actually delivers premiums, and as you state it is hard to really say anything too concrete. I guess part of my question was more along the lines of is it possible for a fund to outperform an index fund based purely on the fact that they trade more patiently? or does the indexing strategy mostly come out ahead anyways since then zero judgement and thereby human error is put in the equation? With that being said it may be hard to find something too concrete regarding the question due to the disclosure requirements
In regards to the poor YTD performance, according to DFA that is due to the value premium not showing up to the same extent. Ofcourse it is possible to argue that is an indicator it may not be as prominent. However basing that purely on the last years data means one one is merely looking at the short term and is within the realm of the active investor. One can ofcourse also argue that the Value premium is found when looking at past results and that past results are not an indiciator of future performance. However there is not much else to go on other than the fact that the premium has been robust when looking back in time and perhaps more importantly the premium has been delivered since its implementation within funds.
Last edited by Flugugrubah on Sun Dec 10, 2017 3:19 pm, edited 1 time in total.
-
- Posts: 21
- Joined: Mon Oct 09, 2017 12:09 am
Re: Do DFA Funds have lower trading costs than Index Funds?
Yeah, that's an interesting point. They did bring up explicit costs of trading and said that their direct market access meant they could have low explicit costs.lack_ey wrote: ↑Sun Dec 10, 2017 2:02 pm Sure, this has been one of the points DFA has made for a long time.
This is one of the advantages to not following an index, really. Decent index fund managers are cognizant of trading costs too, but they have a mandate on performance (keeping transaction costs low) as well as tracking consistency with respect to the index, which they thus have to trade around in an environment where other traders know what they're up to. If you care less about maintaining your ideal portfolio* and more about trading frictions, then you can reduce trading frictions.
*for DFA's case, this may mean slightly lower factor exposures, not having exactly the set of stocks that are being targeted, though as you mention here they downplay this and say that many stocks are available that fit the criteria, which is largely probably true
DFA is not the only one doing this, but it should be an advantage over index funds (at the expense of less transparency, not having a public index to benchmark the fund against, etc.) and certainly over many active funds. If a discretionary active manager wants to buy stock X or really wants to sell stock Y, they may want to move more aggressively at that moment, at the expense of effectively eating higher transaction costs. The active manager would say that their judgment of the timing is more important than that additional cost for more timely execution, but if you generally don't believe in their judgement of the timing, then from that perspective they're just eating higher costs.
So yeah, DFA can have lower implicit costs from trading. They just conveniently have higher explicit costs than many index funds, offsetting those implicit costs and more, potentially. I guess it would depend on the exact trading costs for the given market targeted, ERs, and so on. Also, in some categories DFA could have higher turnover to deal with. Even if more efficient trading per dollar, the net effect could be larger if there are more dollars being turned over.
That was a point I brought up too, that a higher turnover rate than the index fund would raise overall the trading costs. In response to that they said they always put a focus on making sure the value generated from the expected premium earned from trading out of a certain stock or into another, was higher than the costs attributed to it.
Re: Do DFA Funds have lower trading costs than Index Funds?
Good thing there are a lot of index funds that use indexes that don't rebalance once a year, like Vanguard, which uses the CRSP indexes, which don't. Looks like it is quarterly.Flugugrubah wrote: ↑Sun Dec 10, 2017 1:51 pm
They stated that index funds in general rebalance once every year when the index they track rebalances. When this occurs the index funds need to shift their allocations accordingly and quickly to minimize tracking error. This leads to non flexibility in regards to time (since it needs to occur quickly) and non flexibility in regards to quantity (since they need to buy exactly what is included in the underlying index they are tracking). Due to this they need to be flexible in regards to the third factor namely price. Meaning that even if many index funds rebalance during the same time period and move the market, they are all still forced to buy, in doing so they need to cross the bid ask spread making them lose out in that regard aswell.
While the market impact costs aren't published anywhere, expense ratios and brokerage commissions are. Go take a look at the Statement of Additional (SAI) for say Vanguard's and DFA's small cap value or large cap value funds. Then let us know who paid more in brokerage commissions.
"It is difficult to get a man to understand something, when his salary depends upon his not understanding it!" - Upton Sinclair
-
- Posts: 21
- Joined: Mon Oct 09, 2017 12:09 am
Re: Do DFA Funds have lower trading costs than Index Funds?
In regards to implicit costs I believe it would only be beneficial for Vanguard to rebalance more than once a year if other funds dont do it then aswell. And it would only be beneficial to the extent that they would be the only market movers at that point in time meaning the effect may be less extreme. Other than that rebalancing more often means more implicit costs for index funds since they have no flexibility in regards to quantity or time as i stated in my initial post, meaning they have to move the market and have to cross bid ask spreads.alec wrote: ↑Sun Dec 10, 2017 3:20 pmGood thing there are a lot of index funds that use indexes that don't rebalance once a year, like Vanguard, which uses the CRSP indexes, which don't. Looks like it is quarterly.Flugugrubah wrote: ↑Sun Dec 10, 2017 1:51 pm
They stated that index funds in general rebalance once every year when the index they track rebalances. When this occurs the index funds need to shift their allocations accordingly and quickly to minimize tracking error. This leads to non flexibility in regards to time (since it needs to occur quickly) and non flexibility in regards to quantity (since they need to buy exactly what is included in the underlying index they are tracking). Due to this they need to be flexible in regards to the third factor namely price. Meaning that even if many index funds rebalance during the same time period and move the market, they are all still forced to buy, in doing so they need to cross the bid ask spread making them lose out in that regard aswell.
While the market impact costs aren't published anywhere, expense ratios and brokerage commissions are. Go take a look at the Statement of Additional (SAI) for say Vanguard's and DFA's small cap value or large cap value funds. Then let us know who paid more in brokerage commissions.
In regards to the explicit costs I think that is a very valid point, and what would be interesting to see is if potentially lower implicit costs from DFA funds could offset the gain reaped by a bigger company such as for example Vanguard due to lower explicit costs through lower brokerage fees (given that they are able to receive lower brokerage fees than DFA)
- Taylor Larimore
- Posts: 32839
- Joined: Tue Feb 27, 2007 7:09 pm
- Location: Miami FL
Re: Do DFA Funds have lower trading costs than Index Funds?
Flugugrubah:Flugugrubah wrote: ↑Sun Dec 10, 2017 1:51 pm I recently attended a DFA foundations conference in which they described their factor based investment strategy
One thing mentioned at the conference which really stood out to me was that they stated the overall trading costs of their funds was lower than both active and index fund counterparts. In regards to active they stated DFA funds have lower turnover rates, which leads to lower overall cost per definition. In regards to index funds they stated that DFA funds had much lower Implicit costs, which they defined as costs related to market moving and costs due to the bid ask spread.
Their explanation for this was as follows:
When trading on the market one can see it like a negotiation between sellers and buyers, in these negotiations there are three main factors that each party puts different levels of priority on. Time, which means how directly you are in need of the asset, Quantity, which they defined as how specific you are regarding what you want, for example do you want one specific small cap stock or are you indifferent between buying 5 or 10 different stocks. And finally price, which they defined as the price you in the end are willing to settle for. They stated you can prioritize two of these factors at a time, meaning you have to be flexible in the third factor and "accept what you get" there.
They stated that index funds in general rebalance once every year when the index they track rebalances. When this occurs the index funds need to shift their allocations accordingly and quickly to minimize tracking error. This leads to non flexibility in regards to time (since it needs to occur quickly) and non flexibility in regards to quantity (since they need to buy exactly what is included in the underlying index they are tracking). Due to this they need to be flexible in regards to the third factor namely price. Meaning that even if many index funds rebalance during the same time period and move the market, they are all still forced to buy, in doing so they need to cross the bid ask spread making them lose out in that regard aswell.
They then stated that since they gave their traders a high degree of flexibility in regards to quantity ie. which specific stock to buy (as long as the stock fulfilled the factor criterias, which alot of stocks tended to do, it was a buy) as well as flexibility in regards to time (since they do not track a specific index and therefore do not need to shift allocations quickly to minimize tracking error). DFA can be less flexible in regards to price, meaning they get more beneficial prices and therefore are able to have lower implicit trading costs. This is possible since they dont ever need to cross the bid ask spread and can instead wait for the price to come to them ie. place an order on the bid and wait for the transaction to go through. They can also buy in smaller quantities over a long time period meaning they do not move the market.
They then showed a study they had conducted which highlighted how they in general buy for a lower price and sell for a higher price than other funds. They also showed how they on average buy at a lower price than the two adjacent trades, and sell for a higher price than the two adjacent trades.
It was very interesting to listen to and I was hoping to get some Bogelheads thoughts on this
All input is greatly appreciated!
"Trading costs" are reflected in a fund's return. I compared the returns of Vanguard 500 Index Fund (VFIAX) and DFAs DFUSX which attempts to track the 500 Index. Their returns are almost identical suggesting that DFA trading strategies are no better than Vanguard's.
What is often missing in these comparisons is that DFA funds require a financial advisor who charges extra (and may be worth it). This makes the DFA funds more expensive.
Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
Re: Do DFA Funds have lower trading costs than Index Funds?
Until these funds are accessible to the general public not thru an advisor and the advisory fee it doesn't really matter.
- nisiprius
- Advisory Board
- Posts: 52107
- Joined: Thu Jul 26, 2007 9:33 am
- Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry
Re: Do DFA Funds have lower trading costs than Index Funds?
(Oops... Taylor beat me to it)
Look, folks, for an index fund we don't actually need to know what the costs are, because we can just compare the fund's actual results to the index.
For 2012, 13, 14, 15, 16, and 2017 YTD the Vanguard 500 Index fund lagged the index by 0.04%, 0.06%, 0.05%, 0.02%, 0.03%, and 0.04%, respectively. Given that the stated expense ratio is 0.04%, that suggests that any transaction costs (possibly involving positive returns from securities lending, I dunno) are as close to zip as makes no difference.
For DFA, since their funds do not track indexes (and their chosen benchmarks are often bizarre), maybe their costs are lower than those of Vanguard index funds, maybe not, but we have no way of knowing except their say-so, and if they don't specifically call out Vanguard they may just be claiming that their transaction costs are lower than for the Rydex S&P 500 Index Fund (the one with the 1.57% expense ratio).
The thing about DFA funds is that many of them have indeed outperformed similar Vanguard funds, but so have many famous active funds--and the active funds typically have outperformed by more. So the "proof of the pudding" argument would actually lead you to use active funds, and you're just sort of left scratching your head over whether DFA's funds are superior index funds or whether they are inferior active funds.
For example, many of use Vanguard Total Stock, as our core stock holding, classified as large blend. The obvious DFA counterpart would be DFEOX, U.S. Core Equity Portfolio; one obvious famous mainstream active fund would be Fidelity Contrafund.
Source
If I am willing to consider the idea that DFA has an edge over Vanguard due to lower trading costs thanks to patient trading, why shouldn't I be willing to consider the idea that Fidelity has an edge over DFA due to Will Danoff?
Look, folks, for an index fund we don't actually need to know what the costs are, because we can just compare the fund's actual results to the index.
For 2012, 13, 14, 15, 16, and 2017 YTD the Vanguard 500 Index fund lagged the index by 0.04%, 0.06%, 0.05%, 0.02%, 0.03%, and 0.04%, respectively. Given that the stated expense ratio is 0.04%, that suggests that any transaction costs (possibly involving positive returns from securities lending, I dunno) are as close to zip as makes no difference.
For DFA, since their funds do not track indexes (and their chosen benchmarks are often bizarre), maybe their costs are lower than those of Vanguard index funds, maybe not, but we have no way of knowing except their say-so, and if they don't specifically call out Vanguard they may just be claiming that their transaction costs are lower than for the Rydex S&P 500 Index Fund (the one with the 1.57% expense ratio).
The thing about DFA funds is that many of them have indeed outperformed similar Vanguard funds, but so have many famous active funds--and the active funds typically have outperformed by more. So the "proof of the pudding" argument would actually lead you to use active funds, and you're just sort of left scratching your head over whether DFA's funds are superior index funds or whether they are inferior active funds.
For example, many of use Vanguard Total Stock, as our core stock holding, classified as large blend. The obvious DFA counterpart would be DFEOX, U.S. Core Equity Portfolio; one obvious famous mainstream active fund would be Fidelity Contrafund.
Source
If I am willing to consider the idea that DFA has an edge over Vanguard due to lower trading costs thanks to patient trading, why shouldn't I be willing to consider the idea that Fidelity has an edge over DFA due to Will Danoff?
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Re: Do DFA Funds have lower trading costs than Index Funds?
If true, do the lower trading costs result in lower costs for their customers?
Re: Do DFA Funds have lower trading costs than Index Funds?
Last edited by lack_ey on Sun Dec 10, 2017 4:29 pm, edited 1 time in total.
-
- Posts: 15289
- Joined: Fri Apr 10, 2015 12:29 am
Re: Do DFA Funds have lower trading costs than Index Funds?
This applies to many value indices, and to all Russell indices. The S&P500 and most or all total market indices do not reconstitute at fixed, predetermined points in time.They stated that index funds in general rebalance once every year when the index they track rebalances. When this occurs the index funds need to shift their allocations accordingly and quickly to minimize tracking error.
- triceratop
- Posts: 5838
- Joined: Tue Aug 04, 2015 8:20 pm
- Location: la la land
Re: Do DFA Funds have lower trading costs than Index Funds?
That is funny in the context of this thread given the trading costs of using, say, the Small Cap JHSC ETF (click "Tradeability"): High spread, typically high premium/discount, low volume (not always important, but telling here if you want to invest any serious amount)
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."
Re: Do DFA Funds have lower trading costs than Index Funds?
Those ETFs appear to be at least twice as expensive as the pending Vanguard factor funds.
- nisiprius
- Advisory Board
- Posts: 52107
- Joined: Thu Jul 26, 2007 9:33 am
- Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry
Re: Do DFA Funds have lower trading costs than Index Funds?
DFA funds are, in fact, low cost by most standards. (Morningstar's, for example, although I think Morningstar is pretty lenient about giving funds a fee level of "low.")
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Re: Do DFA Funds have lower trading costs than Index Funds?
My understanding is that the fund companies that do indexing are very much aware of the front running that traders do to index funds and that they have themselves adopted patient trading strategies. Is there a Vanguard expert out there who can verify this?
A fool and his money are good for business.
- nisiprius
- Advisory Board
- Posts: 52107
- Joined: Thu Jul 26, 2007 9:33 am
- Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry
Re: Do DFA Funds have lower trading costs than Index Funds?
In point of fact the construction of indexes themselves has taken this into account. It was a feature of the indexes Vanguard co-designed with MSCI and it is a feature of the indexes designed by CRSP. For example, the date at which a stock is added to the index is slightly randomized. Some random verbiage from CRSP's Equity Indexes Methodology Guide
It also contains rules to avoid stocks from quickly shifting in and out of an index (and triggering excessive trading in an index fund):CRSP uses the closing price for all eligible securities on a “random price day” to determine the company’s total market capitalization used at ranking. The random price day is selected algorithmically from the seven trading days immediately prior to the ranking day according to the following convention:
▪ The two days with the highest aggregate absolute price moves (using the CRSP U.S. Total Market Index as the point of reference) are excluded.
PAGE 10
▪ The two days with the lowest aggregate trading volume (using the CRSP U.S. Total Market Index as the point of reference) are excluded.
▪ The excluded days above may overlap.
▪ The remaining days will be used to randomly select the random price day.
In order to prevent premature movement of companies near established breakpoints, CRSP considers bands around each breakpoint, which cushions the security’s movement between indexes through the idea of packeting. A packet is de ned as 50 percent of the index’s total holdings. Due to migration rules, securities may be held simultaneously in multiple indexes.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Re: Do DFA Funds have lower trading costs than Index Funds?
The index front-running issue can be mitigated but maybe not eliminated. This affects the indexes themselves, too, not just the performance of funds relative to the index, as stocks are traded up or down prior to the index changing.
You can predict which stocks are likely going to switch indexes, just maybe with less certainty for certain index providers depending on the methodology.
An index fund manager can't be so patient on trading that the fund's portfolio differs materially from the index for weeks to avoid the price dislocations. Somebody like DFA (or let's say, Vanguard with their upcoming factor ETFs if they want to) doesn't have this concern and can wait to be a liquidity provider or at least not make much market impact.
You can predict which stocks are likely going to switch indexes, just maybe with less certainty for certain index providers depending on the methodology.
An index fund manager can't be so patient on trading that the fund's portfolio differs materially from the index for weeks to avoid the price dislocations. Somebody like DFA (or let's say, Vanguard with their upcoming factor ETFs if they want to) doesn't have this concern and can wait to be a liquidity provider or at least not make much market impact.
Re: Do DFA Funds have lower trading costs than Index Funds?
.
While the message from DFA (in the OP) sounds reasonable, the key question to ask is how significant is it. Or rather, I would pose the question this way - do DFA trading strategies significantly lower costs of simply change factor exposure?
This earlier paper by Donald Keim on the DFA Microcap fund suggest the latter. http://pillarcapital.com/acorn/themes/p ... _paper.pdf
From the paper:
Significance levels (t-stats) on impact of trading strategies
Alpha = 0.15
Mkt = 0.56
SmB = 2.66
HmL = -1.28
Trading strategies significantly increased small cap exposure, with the 'alpha' impact (which embodies trading costs) being statistically indistinguishable from zero.
As others have pointed out, what matters are overall costs (not just trading costs), and I would add factor exposure. As Bill Bernstein said "Over the long-haul what matters is factor exposure and expense."
Posted earlier - here's a comparison of two ex-ante “factor matched” portfolios – portfolio 1 = non-DFA funds, portfolio 2 = DFA funds. These portfolios had the same estimated factor exposure (tilts to market, size, and value), calculated in 2008.
See much of a difference in performance (from trading strategies + other things we are told DFA does)?
Would just note that these comparisons are without any advisor fee subtracted from the portfolio of DFA funds.
https://www.portfoliovisualizer.com/bac ... ion14_2=25
Robert
.
While the message from DFA (in the OP) sounds reasonable, the key question to ask is how significant is it. Or rather, I would pose the question this way - do DFA trading strategies significantly lower costs of simply change factor exposure?
This earlier paper by Donald Keim on the DFA Microcap fund suggest the latter. http://pillarcapital.com/acorn/themes/p ... _paper.pdf
From the paper:
Significance levels (t-stats) on impact of trading strategies
Alpha = 0.15
Mkt = 0.56
SmB = 2.66
HmL = -1.28
Trading strategies significantly increased small cap exposure, with the 'alpha' impact (which embodies trading costs) being statistically indistinguishable from zero.
As others have pointed out, what matters are overall costs (not just trading costs), and I would add factor exposure. As Bill Bernstein said "Over the long-haul what matters is factor exposure and expense."
Posted earlier - here's a comparison of two ex-ante “factor matched” portfolios – portfolio 1 = non-DFA funds, portfolio 2 = DFA funds. These portfolios had the same estimated factor exposure (tilts to market, size, and value), calculated in 2008.
See much of a difference in performance (from trading strategies + other things we are told DFA does)?
Would just note that these comparisons are without any advisor fee subtracted from the portfolio of DFA funds.
https://www.portfoliovisualizer.com/bac ... ion14_2=25
Robert
.
- SmileyFace
- Posts: 9081
- Joined: Wed Feb 19, 2014 9:11 am
Re: Do DFA Funds have lower trading costs than Index Funds?
This isn't necessarily true. I have direct access to DFA funds in my 401K plan and there are no financial advisor costs.Taylor Larimore wrote: ↑Sun Dec 10, 2017 4:00 pm What is often missing in these comparisons is that DFA funds require a financial advisor who charges extra (and may be worth it). This makes the DFA funds more expensive.
Re: Do DFA Funds have lower trading costs than Index Funds?
That is only true for 401k plans. The retail investor has to use a financial advisor. I am not sold on their methodology as I have asked them several times over which time periods they expect outperformance and I never really get a good answer.
-
- Posts: 5561
- Joined: Fri Feb 23, 2007 7:21 pm
Re: Do DFA Funds have lower trading costs than Index Funds?
I would run if I got a specific answer on that one. There are no guarantees in investing. Maybe your not so good answers reflect some integrity on the part of advisers? Check out Larry Swedroe’s Factor Investing book to get a sense of time frames and factors.I am not sold on their methodology as I have asked them several times over which time periods they expect outperformance and I never really get a good answer.
Dave
Re: Do DFA Funds have lower trading costs than Index Funds?
My question was in the context of being shown their results for several funds over many different time periods where their trading, etc. should help but I didn't so I put most of it down to randomness. I understand the advantages of their approach, I guess I just wanted to be shown where it worked without using some long period of back tested results.
- Taylor Larimore
- Posts: 32839
- Joined: Tue Feb 27, 2007 7:09 pm
- Location: Miami FL
Re: Do DFA Funds have lower trading costs than Index Funds?
Draft Investor:DaftInvestor wrote: ↑Mon Dec 11, 2017 1:22 pmThis isn't necessarily true. I have direct access to DFA funds in my 401K plan and there are no financial advisor costs.Taylor Larimore wrote: ↑Sun Dec 10, 2017 4:00 pm What is often missing in these comparisons is that DFA funds require a financial advisor who charges extra (and may be worth it). This makes the DFA funds more expensive.
We may be struggling with definitions, but read this:
9 Things You Need To Know About 401k Fees
Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
- SmileyFace
- Posts: 9081
- Joined: Wed Feb 19, 2014 9:11 am
Re: Do DFA Funds have lower trading costs than Index Funds?
Thanks for the article. To be more accurate - perhaps I should have stated "It costs me no more than the difference in Expense Ratio to hold a DFA fund versus other funds offered to me in my 401K" - there is no financial advisor costs in my 401k being paid by myself.Taylor Larimore wrote: ↑Mon Dec 11, 2017 4:19 pmDraft Investor:DaftInvestor wrote: ↑Mon Dec 11, 2017 1:22 pmThis isn't necessarily true. I have direct access to DFA funds in my 401K plan and there are no financial advisor costs.Taylor Larimore wrote: ↑Sun Dec 10, 2017 4:00 pm What is often missing in these comparisons is that DFA funds require a financial advisor who charges extra (and may be worth it). This makes the DFA funds more expensive.
We may be struggling with definitions, but read this:
9 Things You Need To Know About 401k Fees
Best wishes.
Taylor
I do review my 401K SPD and SAR yearly - in my case - expenses not covered by ERs are covered by employer and they list their costs and I see no financial advisor costs listed anywhere (paid out by myself or employer). Unless I am missing something - the only difference I pay for holding a DFA fund versus other funds in my 401K is the difference in Expense-Ratio (and actually - the DFA expense ratio for the international fund I am holding is less than the managed fund alternative my plan offers).
-
- Posts: 15289
- Joined: Fri Apr 10, 2015 12:29 am
Re: Do DFA Funds have lower trading costs than Index Funds?
It is not hard to look up the ER of DFA funds on the DFA website and compare to what is offered in a 401K. Plans I'm familiar with that include DFA funds range from 0 to 12 bp/year over DFA base cost.