How accurate are the cost savings in Vanguard's Portfolio Watch?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
A440
Posts: 405
Joined: Mon Sep 02, 2013 7:46 am
Location: NJ

How accurate are the cost savings in Vanguard's Portfolio Watch?

Post by A440 » Thu Jul 11, 2019 5:16 am

My Vanguard Portfolio Watch shows a cost savings of close to $11,000 annually when compared to the industry average. 97% of my portfolio are index funds (3-fund model).
I mentioned this statistic to one of our 403(b)(7) reps (one of the better ones) in my school teachers lounge and he thought it was too high. So just how accurate is the cost savings statistic in Portfolio Watch?
I don't know what the future holds, but I know who holds my future.

livesoft
Posts: 67004
Joined: Thu Mar 01, 2007 8:00 pm

Re: How accurate are the cost savings in Vanguard's Portfolio Watch?

Post by livesoft » Thu Jul 11, 2019 6:31 am

Does it matter really? The tool tells you how it was calculated and certainly expense ratios have dropped in the past 10 years, so the Vanguard estimate has to be higher than what is the average today. That written, you can easily calculate your actual cost savings yourself since you know all the expense ratios that are used for that calculation.

I'm sorry to read that a 403(b) rep had to go into a teachers lounge.
Wiki This signature message sponsored by sscritic: Learn to fish.

User avatar
nisiprius
Advisory Board
Posts: 38510
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: How accurate are the costs savings in Vanguard's Portfolio Watch?

Post by nisiprius » Thu Jul 11, 2019 6:47 am

I started to write a post defending Vanguard, but on looking into it some more, I think yeah, Vanguard put a thumb on the scales. Shame on them. Vanguard compares everything to one all-purpose 1.02% "industry average." On looking at some material from the Investment Company Institute my snap judgement is that 1.02% is not defensible. If you have to pick one expense ratio to compare everyone's portfolio to, you want something from the "asset-weighted average" column of figure 6.5 below. 0.40% is defensible ("most portfolios are like a target-date fund.") 0.66% is defensible ("most portfolios are mixtures of stocks and bonds.") 1.02% IMHO is not.

But unless your rep is talking about numbers it's hard to get serious about it. An off-the-cuff "Whaaaaat? It can't possibly be that big, it just can't be" is wrong.

A big issue is how you predict future returns, since total dollar cost savings are highly sensitive to this assumption.

I think the easiest mental arithmetic method to appreciate expense ratios is to calculate the expense ratio as a percentage of assumed return. For example, Fidelity Freedom 2055 returned about 7.08% over the last five years, you can get that from those 1-3-5-10-year return tables your rep probably has. Without expenses, it would have been 7.83%. Divide 0.75% by 7.83%, and you get expenses = 9.57% of returns, call it 10%. That's the percentage you should think about, not the tiny-sounding 0.75% raw expense ratio. Fidelity is taking 10% off the top of every dollar it adds to your retirement account balance.

Some experiments show that Portfolio Watch always compares my portfolio to the same number, a 1.02% "industry average," presumably for all mutual funds of all kinds. The Portfolio Watch results web page says
Note: Expense ratio data provided by Vanguard and Lipper Inc. as of December 31, 2014.
That isn't a crazy number. But, since expense ratios have been coming down--probably because of competitive pressure from Vanguard!--2014 is a little out of date. A more serious issue is that the Investment Company Institute's ICI Fact Book 2019 shows a big difference between "simple average expense ratio" and "asset-weighted average expense ratio." As one would hope, the bigger and more popular funds have lower expense ratios. Both of these are illustrated in Figure 6.3:

FIGURE 6.3
Fund Shareholders Paid Below-Average Expense Ratios for Equity Mutual Funds


Image

I think the "asset-weighted average" is more appropriate. The ICI has several charts suggesting that as one would hope, investors put more dollars into lower-cost funds. The funds you tend to see in 403(b)(7) plans tend to avoid really high cost funds. There's a good relationship between really high expenses and low Morningstar star ratings, so even the dumb but common practice of picking only funds with high star ratings will tend to screen out really high-cost funds.

Another question of whether it's fair to compare your investments with an industry average, because anyone with the gumption to get as far as an expense comparison tool has the gumption not to buy an S&P 500 index fund with a 1.58% expense ratio. (That's not a joke, that's RYSOX).

I think the most relevant single table in the ICI report is figure 6.5. (They explain their term "hybrid" on p. 76: "Hybrid funds (also called asset allocation funds or balanced funds)"

Image

So, what to use as a single number? Definitely, an asset-weighted average. And, definitely, a number more recent than 2014.

Since most of us invest in a mixture of stocks and bonds, I think the "hybrid" number, 0.66% would be defensible. On the other hand, it's pretty noticeable that the extremely popular target date funds are now down to an average expense ratio of only 0.40%. Since such funds are widely available in employer retirement plans, you can make out a good case that if your personal portfolio is not too different from a target-date retirement fund, the appropriate expense comparison would be with the (asset-weighted) average target-date retirement fund.
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.

livesoft
Posts: 67004
Joined: Thu Mar 01, 2007 8:00 pm

Re: How accurate are the cost savings in Vanguard's Portfolio Watch?

Post by livesoft » Thu Jul 11, 2019 7:24 am

And apparently not included in the analysis are the effects of front-end loads which many 403(b) plans have.
Wiki This signature message sponsored by sscritic: Learn to fish.

Topic Author
A440
Posts: 405
Joined: Mon Sep 02, 2013 7:46 am
Location: NJ

Re: How accurate are the cost savings in Vanguard's Portfolio Watch?

Post by A440 » Thu Jul 11, 2019 8:27 am

livesoft wrote:
Thu Jul 11, 2019 6:31 am
I'm sorry to read that a 403(b) rep had to go into a teachers lounge.
In our district, this is the only place in the school they are allowed. I often find the need to "escort" some AXA/Equitable reps out of teacher's rooms and other places they shouldn't be to places they should be. It's usually the "newbies" from this company. The other reps seem know the rules. :happy
I don't know what the future holds, but I know who holds my future.

bloom2708
Posts: 6392
Joined: Wed Apr 02, 2014 2:08 pm
Location: Fargo, ND

Re: How accurate are the cost savings in Vanguard's Portfolio Watch?

Post by bloom2708 » Thu Jul 11, 2019 8:39 am

A440 wrote:
Thu Jul 11, 2019 5:16 am
My Vanguard Portfolio Watch shows a cost savings of close to $11,000 annually when compared to the industry average. 97% of my portfolio are index funds (3-fund model).
I mentioned this statistic to one of our 403(b)(7) reps (one of the better ones) in my school teachers lounge and he thought it was too high. So just how accurate is the cost savings statistic in Portfolio Watch?
https://www.dinkytown.net/java/compare- ... -fees.html

You can easily buy funds with expense ratios of 1, 1.5% and 2%+. Maybe not from Vanguard, but they are out there.

My (short term) EJ guy had me in an Oppenheimer fund with a 2.35% expense ratio.

Play around with the calculator above and try different expense ratio levels. See the impact over 20, 30, 40 years.
"We are not here to agree with you; we are here to provoke thoughtfulness." Unknown Boglehead

Post Reply