You might know the expense ratio of the mutual funds you own, but do you know how much of your money they spend on brokerage commissions? Yes, even huge billion-dollar fund managers have to pay someone to execute their stock trades for them, and it can sometimes involve a murky world of “soft” dollars where expenses are hidden from investors through inflated stock commissions.

Morningstar article A Big Fund Cost You Don’t See explored this hidden cost and found that the average equity fund pays approximately 0.30% of assets a year in brokerage commissions. If your equity reports an expense ratio of 0.90%, that would bring the total expenses including commissions to 1.20%. (When comparing performance numbers, those are usually reported after all expenses.)

However, these are just averages. Some funds have high turnover and thus lots of trades relative to their asset balances. The Morningstar article highlighted 10 funds that paid over 0.60% of assets in brokerage commissions recently. One of these is CGM Focus (CGMFX), which I seem to read about a lot in popular personal finance magazines. They spend 1.04% of assets on stock commissions, on top of the 1.23% official expense ratio. Overcoming a 2.27% total expense drag every single year is quite difficult – some would say nearly impossible over the long run.

I am not saying they do this, but you can see how it would be beneficial for a mutual fund to shift some of their costs onto the stock commission side and reduce the highly publicized expense ratio figure.

I also ran across this Mutual Funds’ Hidden Costs page at Retire Early Home Page, which was last updated in 2005 but provides some historical perspective. In 2004, Forbes magazine revealed that the Dreyfus Founders Passport Fund paid an “astonishing” 4.64% of assets towards brokerage commissions.

It also states (again back in 2004):

The average US domestic mutual fund pays about 5 cents in brokerage commissions for every share of stock bought or sold in the fund. Vanguard averages about 2 cents per share in its low-cost index funds.

What about Vanguard index funds?
This made me want to investigate Vanguard’s brokerage commission hit further. To calculate it yourself for any mutual fund, you’ll need two numbers. First, you’ll need the total net assets as of December 31st, found inside the fund prospectus. Second, you’ll need to track down the total brokerage commissions spent, found inside the fund Statement of Additional Information (SAI). Luckily, these are both easily found on Vanguard.com for any specific fund – just click on “View prospectus and reports”.

In 2004, the Vanguard Total Stock Market Index – Investor Shares (VTSMX) had an expense ratio of 0.19%, and brokerage commissions of only 0.0077% of net assets In 2008, the expense ratio was 0.18%, with the commission costs dropping to only 0.0069% ($5,643,000 divided by $82 billion). As the fund grows, the expenses per dollar are reduced – something I’ve come to see and expect from Vanguard.

This goes back to one of the structural features of index funds – they do not need to trade very much to maintain their goal of tracking an index based on market capitalization. If one company rises (or drops) even significantly in value, they simply represent more (or less) of the index accordingly. There is little need to buy or sell in response to market fluctuations. Instead, trades are made primarily due to new fund purchases and redemptions by investors, or due to the occasional change in the tracking index.

Have you seen those “follow the green line” ads from Fidelity? Well, they reminded that a reader sent me their retirement account holdings for review which was managed through the Fidelity Portfolio Advisory Service (PAS). This is a managed portfolio service, which means that you pay Fidelity a fee and they do all the research, selection, buying, and selling for you. Fidelity has two managed-portfolio tiers for individual investors, with the Portfolio Advisory Service for account balances of $50,000+, and the Private Portfolio Service for those with $300,000+ to invest.

At only a $50,000 minimum portfolio size, it appears that the PAS is targeted a relatively large portion of the generic public. Unfortunately, in the wealth management business such small balances usually also mean generic, cookie-cutter portfolios with little or no personalization. Here’s what Fidelity says:

In the Fidelity Portfolio Advisory Service product, customers are invested into model portfolios of Fidelity and non-Fidelity mutual funds based on their time horizon, risk tolerance and investment goals. These model portfolios are managed by a team of investment professionals that includes Portfolio Strategists and Mutual Fund Analysts.

Sounds like “we make you answer a short questionnaire and the computer spits out an asset allocation” to me. Let’s see how Fidelity constructs this person’s portfolio. I will mention here that this is an IRA account, so that taxes aren’t a huge concern.

Portfolio Comparisons

Benchmark Portfolio
This particular account used the “Growth w/ Income Portfolio” benchmark, which has an overall 60% Stocks/40% Bonds balance. Other portfolio options are Conservative (20% stocks), Balanced (50% stocks), Growth (70% stocks), Aggressive Growth (85% stocks), and All Equity (100% stocks). Here are the indexes and asset allocation for this portfolio.

Stock
50% Total US (Dow Jones US Total Stock Market Index)
10% Developed International (MSCI EAFA Index)

Bond
25% US Investment Grade Bond (Barclays Capital US Aggregate Bond Index)
10% US High-Yield (Merrill Lynch US High Yield Master II Constrained Index)

Cash
5% Treasury Bills (Barclays Capital 3-month US T-Bill Index)

Hypothetical Index Fund Portfolio
As mentioned in the statement, you cannot invest in an index. So, I created below a portfolio consisting of actual investments that passively track the above indexes with minimal costs. I chose the cheapest ETF that tracks the exact index if possible, not the cheapest ETF that was similar. I also included the annual expense ratios.

50% SPDR Dow Jones Total Market ETF (TMW) 0.21%
10% Vanguard Europe Pacific ETF (VEA) 0.16%
25% Vanguard Total Bond Market ETF (BND) 0.14%
10% SPDR Barclays Capital High Yield Bond (JNK) 0.40%
5% SPDR Barclays Capital 1-3 Month T-Bill (BIL) 0.13%

The total weighted expense ratio was 0.20%.

Actual Fidelity-Managed Portfolio
The actual choice of investments in this account matches the benchmark asset allocation closely, and included over 30 different mutual funds. You can view the entire mutual fund list here, but here is the overall breakdown:

51.5% US Stock Funds
10.0% International Stock
28.5% Investment-Grade Bonds
10.0% High-Yield Bonds
0% Cash

This includes a mix of twenty (!) different actively-managed domestic stock funds from both Fidelity and outside companies like Janus and T. Rowe Price (Okay, 1.5% was in one index fund – S&P 500 Fidelity Spartan.) The average expense ratio for these was approximately 1%. Six different international stock funds were included, with an average expense ratio of ~1.2%. The overall bond fund expense ratios were 0.8%. This brought the total weighted expense ratio to ~0.94%.

Performance Comparisons

Now for the important part, returns after all fees. This account was not ten years old, so the best long-term return number was the 5-year historical annualized returns. The statement was as of 6/30/09.

First up, we have the 5-year annualized of the Benchmark Portfolio, which as of 6/30/09 was 1.6%. Of course this is a benchmark, which doesn’t include any management fees or commissions.

I was unable to find performance numbers as of 6/30/09 for my Hypothetical Index Fund Portfolio as it is already 2010 (if someone knows how to do this please let me know). However, we can estimate the return since the total weighted expense ratio was 0.20%. If we estimate trade commissions to be $5 per trade x 5 ETFs = $25 per month… on a $100,000 portfolio that is 0.30%. (Such trade commissions would be zero if held at Zecco or WellsTrade, given the account size.) Assuming the ETFs follow the indexes closely, then the 5-year returns would be in the neighborhood of 1.1%.

Now, what was the actual 5-year annualized return on this fully-managed account? -0.6%. Yes, negative 0.6%.

Conclusion

Over the past 5 years, the Fidelity Portfolio Advisory Service managed to construct a portfolio that lagged a simple index fund portfolio by 1.70% annually. That’s a huge difference over time. Use any compound interest calculator and stick in two numbers that differ by 1.7%, and you’ll see the effect of compound interest working against you for a few decades. Why did this account perform so poorly relative to its benchmark? Isn’t it supposed to beat the benchmark?

Too many advisors. To me, if people choose to hire someone to manage their investments, it would be to tap into their expertise and special insight. I’d want him/her to make calculated bets that will beat the market. Putting my money in 34 different mutual funds, each with their own team of advisors, seems like everyone’s bets would cancel each other out.

While the reason given for so many funds was “diversification”, the only phrases that came to my mind were “overlap” and “lack of focus”. You don’t need to own a ton of funds to get diversification. With so many funds, you’d probably end up owning the same companies as the index fund anyway.

Costs matter. The actively-managed mutual funds are the first layer of expenses, which was a weighted 0.94%. Then there is the second layer of management fees charged by Fidelity, which includes all trade commissions and varies from .25%–1.7% based on asset levels. In this account, it was 0.8%. Thus, in order to simply match the benchmark, the investments chosen would need to outperform it by 1.74%. Every. Single. Year. That is a stiff headwind.

The really sad thing is, even if I just invested in the index funds through the Fidelity PAS and basically paid them to do nothing, I would have still done better than the funds they chose. 1.6% index – 0.2% index fund expenses – 0.8% Fidelity fee = gaining 0.60% a year. Compare that with losing 0.60% a year.

For a $200,000 portfolio paying 1.38% in portfolio management fees, that’s $2,760 a year. Don’t pay nearly 3 grand a year for a cookie-cutter asset allocation that doesn’t even match an index fund. It can be well worth your time to learn more about investments yourself. Here are some starting ideas.

One of the hottest asset classes in 2009 by far was Emerging Markets, which includes stocks from developing markets like China, Brazil, Korea, Taiwan, India, and South Africa. Two of the most popular ETFs in this category are the iShares MSCI Emerging Markets Index (EEM) and the Vanguard Emerging Markets ETF (VWO).

Which one should you choose?

Similarities. Both are primarily passively-managed and based on the MSCI Emerging Markets Index. EEM has been around longer but both currently have similar levels of assets, with $40B for EEM and $34B for VWO.

Differences. EEM is less expensive than most other actively-managed Emerging Markets mutual funds at a 0.72% annual expense ratio, but Vanguard is significantly cheaper still at 0.27%. EEM has a much higher average trading volume, as it has a higher popularity amongst institutional investors and active traders.

While they track the same index, both ETFs have their own sampling methods to replicate that index. EEM holds 439 stocks with an average market cap of $23.1 million, while VWO holds 816 stocks with an average market cap of $17.6 million. Unfortunately, Morningstar chose the MSCI EAFA Index as their benchmark for statistical comparison, so it’s hard to see exactly historically which one has been following the index more closely.

Performance-wise, iShares’ EEM has recently been lagging behind Vanguard, as was pointed out in this IndexUniverse post. In 2009, EEM rose 68.8% but VWO gained 76.3%, nearly 8 percentage points more.

This has led to a significant outflow of funds recently from EEM into VWO:

However, if we look back to 2008 EEM dropped 48.9% while VWO dropped 52.5%. VWO isn’t old enough to compare 5- or 10-year historical returns, but if you compare 3-year annualized trailing returns VWO only beats EEM slightly at 2.69% vs. 2.60%. That’s actually surprising given the 45 basis point expense edge for VWO.

Conclusion. Due to their differences, either EEM or VWO may outperform each other in the short run. However, historically the significantly lower expense ratio of VWO has won out, and I would expect it continue to do so in the long run. VWO also holds more stocks and thus would seem to track the MSCI index more closely, while EEM offers more liquidity if you’re trading high amounts of shares at once.

Disclosure: I am or will be invested in VWO or the mutual fund equivalent (VEIEX). The reason I started this research is because I am currently trying to figure out how to switch to an ETF since the Vanguard fund VEIEX is both more expensive at 0.39% annually and has both a purchase fee of 0.50% and a redemption fee of 0.25%. Ouch! You can compare the total ongoing costs of VWO vs. VEIEX easily at this Vanguard calculator.

In addition to their new $7.95 per trade price structure, Fidelity is also waiving commissions completely for 25 iShares ETFs. This appears to be a partnership where iShares partially compensates Fidelity themselves for promoting their ETFs.

Fidelity receives compensation from the ETF sponsor and/or its affiliates in connection with a marketing program that includes promotion of iShares ETFs and certain commission waivers. Additional information about the sources, amounts, and terms of compensation is described in the ETF’s prospectus and related documents. Fidelity may add or waive commissions on ETFs without prior notice.

Technically, you could make a relatively complete passive portfolio consisting entirely of these iShares ETFs. For example, if you wanted a stock portfolio encompassing the entire world in proportion to their current market capitalization share, you could go with

45% IWV (Broad United States)
40% EFA (Broad Developed Europe & Pacific)
15% EEM (Emerging Markets)

Since these ETFs have pretty good trading volume, the bid-ask spreads should also be reasonably small. For the bond portion, one possibility would be

50% AGG (Broad, Investment Grade Taxable US Bonds)
50% TIP (US Treasury Inflation-Protected Bonds)

assuming you had room in a tax-advantaged account like an IRA. Not a recommendation, just what I might buy for myself.

US Equity
iShares Russell 1000 Growth (IWF)
iShares Russell 1000 (IWB)
iShares Russell 1000 Value (IWD)
iShares Russell 2000 Growth (IWO)
iShares Russell 2000 (IWM)
iShares Russell 2000 Value (IWN)
iShares Russell 3000 (IWV)
iShares S&P 500 Growth (IVW)
iShares S&P 500 (IVV)
iShares S&P 500 Value (IVE)
iShares S&P Mid Cap 400 Growth (IJK)
iShares S&P Mid Cap 400 (IJH)
iShares S&P Mid Cap 400 Value (IJJ)
iShares S&P Small Cap 600 Growth (IJT)
iShares S&P Small Cap 600 (IJR)
iShares S&P Small Cap 600 Value (IJS)

Dow Jones U.S. Real Estate (IYR)

International Equity
iShares MSCI ACWI (ACWI)
iShares MSCI EAFE
iShares MSCI EAFE Small Cap (SCZ)
iShares MSCI Emerging Markets

Bonds
iShares Core Total U.S. Bond Market ETF
iShares Barclays TIPS Bond ETF
iBoxx $ Investment Grade Corporate (LQD)
JP Morgan USD Emerging Markets (EMB)
S&P National AMT-Free Municipal (MUB)

Fidelity announced today that starting February 3rd, 2010, the price of all online equity trades through their brokerage would cost $7.95, with no limit on the number of shares. Previously, the commissions ranged from $8 to $19.95 per trade depending on asset value and trading volume. This is great news for my Self-Employed 401k at Fidelity. Phone and broker-assisted trades cost a little more:

Customers placing trades through Fidelity’s Automated Service Telephone (FAST) will pay an additional $5 per trade ($12.95 total), and those placing trades through a representative will pay an additional $25 ($32.95 total) per trade.

This move makes Fidelity more competitive with the other discount brokers out there, for example I have an account with OptionsHouse which charges only $2.95 per trade.

They also announced commission-free trades for 25 iShares ETFs, which seems like a direct response to Schwab’s commission-free ETFs.

Since I am expanding my portfolio to beyond just IRAs and 401ks, I will need to move my more tax-inefficient investments like bonds into those tax-deferred accounts. This way, I’ll put more of my stocks in a plain taxable account which is taxed at the more favorable long-term capital gains rates.

Fidelity is my Solo 401k provider, so I was looking through their bond offerings. Their only bond index funds are the Fidelity U.S. Bond Index Fund (FBIDX) and essentially a US Treasury bond index fund with different maturity lengths. Looking at FBIDX, its benchmark is the Barclays Capital U.S. Aggregate Bond Index. This index covers the all US bonds on the market that are investment grade and taxable, including Treasuries, corporate bonds, and mortgage-backed and asset-backed securities.

This is basically the same benchmark index as the Vanguard Total Bond Market Index fund (VBMFX), which follows the Barclays Capital U.S. Aggregate Float Adjusted Bond Index. (ETF version is ticker BND.) The “float-adjusted” part just accounts for all those Fannie Mae bonds that are now owned by the government and thus not actually available on the market anymore. They should be the same, right? Nope.

A perk of being a financial blogger is companies give you free stuff to try, and Morningstar offered me a free 1-year subscription to their paid Morningstar Premium service which offers stock and mutual fund research. Being a mostly a low-cost index fund investor, most of their content didn’t really interest me. However, I started reading the FBIDX Analyst Report and was surprised. The teaser line was:

Although this fund is designed to track the Barclays Capital U.S. Aggregate Bond Index, it lagged that bogy by more than 3 percentage points from mid-2007 through the end of 2008.

Lagging a benchmark by 3 full percentage points is a lot in the index world. Passive investors want low costs and low tracking error from the target benchmark! That’s how we get our greater-than-average return over the long run. It turns out that Fidelity had been making “modest” bets outside of the index in order to juice their returns, including minor exposure to subprime mortgages through an “ultra-short” bond portfolio*. FBIDX costs more than VBMFX by about 0.12% a year, so they would always lag if all other things were equal, and needed to take such additional risks in order to have better performance numbers.

The report goes on to report that the fund hired a new manager in February 2009, who says he’ll run a tighter ship:

Since joining the fund in February 2009, he [lead manager Curt Hollingsworth] has taken steps to ensure that it will closely track its benchmark from month to month. He thinks actively run internal bond portfolios have no place in this passively run offering, which should reduce the chance of things going awry. He also pared the fund’s exposure to out-of-index bonds from 8% to just 2% of assets lately.

Well, that’s nice to hear, but too late for me. When I see “index” in the fund name, I expect it to match the benchmark, not try to beat it like everyone else. Thus, I’m most likely going to construct my portfolio out of Vanguard bond ETFs instead. My transaction size should be great enough to counteract the hit from each individual stock trade. (Read about the Vanguard ETF vs. mutual fund cost comparison tool.)