Are Trading Costs Included in Expense Ratio?

[Call_Me_Op]

I have read conflicting information on whether a mutual fund's expense ratio includes trading costs. Can someone shed light on this question? I would think it does not, because you cannot predict trading costs ahead of time. A corollary question is whether performance charts typically plot return after all fees and expenses (including trading costs).

[johnep]

Fund expenses do not include trading costs. A high turnover fund will incur a lot of trading costs. Index funds usually have very low turnover, often 3% or less which gives them another cost advantage over active funds. I do not know for sure the answer to your second question, but I think most performance charts would be net of any trading costs. However, I would cautious of those provided by non neutral party, i.e. the fund company promoting their fund.

[porcupine]

I have read conflicting information on whether a mutual fund's expense ratio includes trading costs. Can someone shed light on this question? I would think it does not, because you cannot predict trading costs ahead of time. A corollary question is whether performance charts typically plot return after all fees and expenses (including trading costs).

For what it is worth, here is a similar thread I started which might be helpful. Though I am still unsure of the answers to my questions, I did not revisit the (my) thread because I am yet to do the recommended homework. If you are able to finish it, I would appreciate it if you could share your findings!

Oh BTW, AFAIK (which is probably not a whole lot, in this situation), ER is not a future indicator; rather, it is a trailing indicator, i.e., based on the ER over the past year.

- Porcupine

[Call_Me_Op]

Porcupine,

This question is of fundamental importance, and I would think every Boglehead would want to know the answer. As I said above, it is logical that expense ratio does not include trading costs, since these costs cannot be known a priori.

[dbr]

Trading costs are not in the ER. Report of trading costs can be found somewhere in the fund annual report and/or report of supplementary information. Clearly such costs are reported after the fact as are actual ER's. ER, however, can be pretty closely estimated for the future. Some people have estimated that trading costs might amount to 1% for every 100% in fund turnover for equity funds.

Fund performance is after all costs and distributions are included. Charts of fund NAV include all costs but distributions are lost.

[porcupine]

Porcupine,

This question is of fundamental importance, and I would think every Boglehead would want to know the answer. As I said above, it is logical that expense ratio does not include trading costs, since these costs cannot be known a priori.

CMO:

I agree with what you say. However, like I said, I was told to do some reading up. I did not complete that assignment. From what little I read, I was still not where I could come back and wrap up my thread/questions.

My thread was much more simpler and quantifiable than yours, which includes open ended questions. What I had done was laid out some numbers and, based on that specific information, asked folks to tell me what the ER, costs etc. would be in that specific scenario. I don't think there was one complete answer among the responses I got - only suggestions on what/where to read up on. I did peruse the documents that I recently received from my mutual fund companies. That did not really help me (yet), probably also because I did not read them really thoroughly - each of those documents had at least 50-100 pages apiece.

- Porcupine

[nisiprius]

No, trading costs are not included in the ER. One of the advantages of an index fund is that you can see how closely it tracks the index it's supposed to follow. You know what it's supposed to do--approximate an ideal cost-free investment in an index. You can't actually tell what the trading costs are, but you can calculate exactly how much it "costs" you to invest in the real-world fund, compared to the ideal of a cost-free investment in the index itself.

The fact that trading costs are not included has sometimes been used dishonestly to suggest that the low expense ratios of our favorite index funds is illusory, and to insinuate that the actual costs are high. But by comparing this with the index we can refute this.

For example--and I am deliberately including both a Vanguard and a non-Vanguard product, because this is a "low-cost index fund" thing, not a Vanguard thing--if we look at, say, the Vanguard 500 Index investor shares (ER 0.18%); the SPY ETF (ER 0.09%); and the S&P 500 itself (ER zero)--total return, dividends reinvested--a 5 year growth chart shows us that $10,000 would have grown to:

Vanguard 500 Index Inv:12,356.31
SPDR S&P 500:12,371.51
S&P 500 TR:12,410.00

Now, you can do the exact math or you trust an approximation; for small changes, you can get the difference in annual growth rate just by dividing the total difference by the number of years, and then dividing by the starting amount ($10,000) and converting to a percentage. In this case, the actual costs to you, compared to an ideal cost-free investment in the S&P 500 index itself, are as follows:

Vanguard 500 Index (VFINX) ($12,410.00 - $$12,356.31) = 53.69; $53.69 / 5 = $10.74 = 0.11%
SPDR S&P 500 (SPY): ($12,410.00 - $12,371.51) = $38.49; $38.49 / 5 = $7.70 = 0.08%

The important thing here is that the costs to you, compared to an ideal cost-free investment in the S&P 500 index itself, are about the same as the expense ratios. A point of secondary interest is that both cases, they are actually a bit less than the expense ratio!

The tracking error calculation doesn't actually tell us what trading costs are, but it does tell us either that the costs are very low, or that the fund managers have been able to make up some of those costs by tiny doses of active management, or both.

Since a cap-weighted index fund does not need to trade in order to restore the weighting, turnover is very low--which we can see for ourselves--and it seems extremely plausible that the actual trading costs are very low.

[Call_Me_Op]

The most important consequence of the fact that trading costs are not included in expense ratio is that the full cost of the actively managed funds is hidden.

[porcupine]

The most important consequence of the fact that trading costs are not included in expense ratio is that the full cost of the actively managed funds is hidden.

But do note that the performance of all funds is still AFTER all these costs are removed. In other words, what you see does appear to be what performance you got (and what fees you paid) for a given (past) period.

- Porcupine

[dumbmoney]

Nisi's answer is correct if you narrowly define trading cost to be brokerage commissions, and all costs incurred by fund inflows/outflows. It does NOT include the market impact cost of index changes, since those costs are included in the "cost free" benchmark (a consequence of index changes being announced in advance, which allows index funds to trade perfectly in sync with the index). In other words, index funds embed some of their trading costs in the index that they track. You can't determine the true total cost by subtracting the index fund return from the index return.

[Call_Me_Op]

The most important consequence of the fact that trading costs are not included in expense ratio is that the full cost of the actively managed funds is hidden.

But do note that the performance of all funds is still AFTER all these costs are removed. In other words, what you see does appear to be what performance you got (and what fees you paid) for a given (past) period.

- Porcupine

Indeed. But unless one in careful, they will miss this significant drag (associated with trading costs) when considering actively managed funds. Of course, most who choose them are probably not very cost conscious anyway.

[dbr]

The fact that trading costs are not included has sometimes been used dishonestly to suggest that the low expense ratios of our favorite index funds is illusory, and to insinuate that the actual costs are high. But by comparing this with the index we can refute this.

Probably more accurately that the costs of high ER active funds are even higher than they seem and that low cost funds also have low trading costs.

[Padlin]

Are the added trading costs reflected in the 1Y, 3Y, etc Performance figures you see listed by Morningstar or some such?

[rkuklinski]

I have consulted http://www.personalfund.com/learnmore.html in the past to estimate the impact of trading costs on a particular fund. Here is their rule of thumb on estimating transactions costs:

on average, a fund with 100% annual turnover gives up nearly 1% in transaction costs. A fund with 25% turnover would give up only a quarter as much. A fund with 300% turnover – three times as much.

Transaction costs are not incorporated in a fund's "total expense ratio." They are taken directly out of shareholder assets.

Registration is free and you can access their estimated trading costs for your favorite funds.

[pkcrafter]

Where did the money go? In addition to the brokerage commissions there are other trading costs that are harder to quantify, such as the spread cost (difference between the bid price and the asked price), market impact cost (change in the price of a security due to the trade being made by the fund), and opportunity cost (change in the price of a security between when the fund decides to make the trade and when the trade completes).

Mark Carhart analyzed this issue and he found that, on average, a fund with 100% annual turnover gives up nearly 1% in transaction costs -- 0.95% to be precise. A fund with 25% turnover would give up only a quarter as much. A fund with 300% turnover -- three times as much. And so on.

As previously said, these costs are extracted from the funds before performance figures are posted and they are not part of the ER, although some advertizing may be in the 12b-1 fee which is part of ER. Internal costs may be hard to see in the short term, but very telling in the long term.

Paul

[porcupine]

I have read conflicting information on whether a mutual fund's expense ratio includes trading costs. Can someone shed light on this question? I would think it does not, because you cannot predict trading costs ahead of time. A corollary question is whether performance charts typically plot return after all fees and expenses (including trading costs).

This is why it is so confusing! Almost everyone responding here has said that expense ratios do not include transaction costs. But here is an article (don't have a clue how trustworthy) which says the opposite.

Edited to add: More information

And another link

- Porcupine

[johnep]

That site is incorrect. Here is a link to Investopedia with their definition:
http://www.investopedia.com/terms/e/exp ... z2JMuWZt5i

[johnep]

One other thing about trading costs. The 1% cost for 100% turnover may be a good rule of thumb but I believe these costs can vary widely depending upon the types of securities being bought and sold. The difference between "ask" and "buy" prices can vary widely. Heavily traded securities like S&P500 companies will have very narrow margins whereas more thinly traded companies like small caps or perhaps some foreign stocks would have much wide margins, hence funds that invest in thinly traded securities will have much higher costs for their trade transactions. Larry Swedroe explained this very well in one of his books (sorry i cannot remember which one).

[rkhusky]

From Vanguard's website:

Still another expense is the cost of trading securities, which includes charges such as brokerage commissions. These costs aren't included in a fund's expense ratio.

https://retirementplansp.vanguard.com/V ... ostOwn.jsf

[rkhusky]

One other thing about trading costs. The 1% cost for 100% turnover may be a good rule of thumb but I believe these costs can vary widely depending upon the types of securities being bought and sold. The difference between "ask" and "buy" prices can vary widely. Heavily traded securities like S&P500 companies will have very narrow margins whereas more thinly traded companies like small caps or perhaps some foreign stocks would have much wide margins, hence funds that invest in thinly traded securities will have much higher costs for their trade transactions. Larry Swedroe explained this very well in one of his books (sorry i cannot remember which one).

This is why some Vanguard funds charge extra for buying/selling shares (e.g. the Int'l small cap fund (VFSVX)).

[sscritic]

The SEC sets the rules. Here are some:

Item 3. Risk/Return Summary: Fee Table
Include the following information, in plain English under rule 421(d) under the Securities Act, after Item 2:
Fees and expenses of the Fund
...
Portfolio Turnover
The Fund pays transaction costs, such as commissions, when it buys and sells securities (or “turns over” its portfolio). A higher portfolio turnover rate may indicate higher transaction costs and may result in higher taxes when Fund shares are held in a taxable account. These costs, which are not reflected in annual fund operating expenses or in the example, affect the Fund’s performance.

Here is a Vanguard prospectus:

The Fund pays transaction costs, such as commissions, when it buys and sells securities (or “turns over” its portfolio). A higher portfolio turnover rate may indicate higher transaction costs and may result in more taxes when Fund shares are held in a taxable account. These costs, which are not reflected in annual fund operating expenses or in the previous expense examples, reduce the Fund’s performance.

Note that Vanguard is not using the exact words provided by the SEC in their example.

[dbr]

I have read conflicting information on whether a mutual fund's expense ratio includes trading costs. Can someone shed light on this question? I would think it does not, because you cannot predict trading costs ahead of time. A corollary question is whether performance charts typically plot return after all fees and expenses (including trading costs).

This is why it is so confusing! Almost everyone responding here has said that expense ratios do not include transaction costs. But here is an article (don't have a clue how trustworthy) which says the opposite.

That article is flat out wrong about brokerage costs.

Edited to add: More information

This article, strangely enough, does not seem to mention trading costs. The fees they do mention are fees directly charged to the investor such as loads. That is an entirely different thing.

And another link

This article does not say that trading costs are in the ER or that they are not but does say they are costs that drag down fund performance.

- Porcupine

[nisiprius]

Note that Vanguard is not using the exact words provided by the SEC in their example.

I was not unable to fail to find any point of difference between the two statements. They looked not unidentical. Did you say "not" when you meant to say "not not?"

[sscritic]

affect is not reduce, or at least I an unable to not find a lack of distinction between the two.

[nisiprius]

affect is not reduce, or at least I an unable to not find a lack of distinction between the two.

And I see Vanguard says "more" where the SEC as "higher."

[porcupine]

affect is not reduce, or at least I an unable to not find a lack of distinction between the two.

And I see Vanguard says "more" where the SEC as "higher."

It took me two more tries to see "more"!!

- Porcupine

[Lumpr]

Note that Vanguard is not using the exact words provided by the SEC in their example.

FWIW - They are not required to use the exact words (only to provide comparable narrative). Arguably "reduce" is more precise than "affect" (something that reduces affects, but so does something that increases).

[dodonnell]

Trading Costs are usually listed in the "Statements of Additional Information" or SAI.

For example, you can find (if you dig) the SAI for Vanguard's S&P 500 Index fund here:
http://edgar.sec.gov/cgi-bin/browse-edgar?company=&match=&CIK=VFINX&owner=exclude&Find=Find+Companies&action=getcompany

You will also learn that Vanguard uses:

-derivatives (Futures contracts both long and short)
-loans out stocks to 3rd parties, and
-purchases other fixed income securities

to slightly enhance their return (and also increase the risk) when compared to the "costless index" which some compared it to in earlier posts.

You cannot do simple math on their tracking error to determine Vanguards' trading expenses - you must look it up. Also, you should decompose their portfolio, determine the increase risk they are adding to the portfolio compared to the benchmark portfolio to understand that "there is no free lunch".

Vanguard characterizes this risk here (emphasis added):

C. Futures Contracts: The fund uses index futures contracts to a limited extent, with the objectives of maintaining full exposure to the stock market, enhancing returns, maintaining liquidity, and minimizing transaction costs. The fund may purchase futures contracts to immediately invest incoming cash in the market, or sell futures in response to cash outflows, thereby simulating a fully invested position in the underlying index while maintaining a cash balance for liquidity. The fund may seek to enhance returns by using futures contracts instead of the underlying securities when futures are believed to be priced more attractively than the underlying securities. The primary risks associated with the use of futures contracts are imperfect correlation between changes in market values of stocks held by the fund and the prices of futures contracts, and the possibility of an illiquid market.

ps. an example of this extra risk is when Lehman failed, the spread between futures and the underlying SP500 blew out ... and this risk appeared ... and it cost 3-4% in additional loss on the futures contracts over the underlying

[sscritic]

Note that Vanguard is not using the exact words provided by the SEC in their example.

FWIW - They are not required to use the exact words (only to provide comparable narrative). Arguably "reduce" is more precise than "affect" (something that reduces affects, but so does something that increases).

I see I was not precise. The SEC provides an example, thus an exact copy is not required. I did use "in their example" and perhaps someone thought that "their" referred to Vanguard. I wouldn't myself, as the immediate antecedent was "the SEC." If I had wanted to say that it was Vanguard's example, I would have written

Note that in their example Vanguard is not using the exact words provided by the SEC.

placing their example next to Vanguard.

Or were you worried people didn't know what example meant and might have interpreted example as exact copy?