Thoughts on improving a lazy ETF portfolio

[FinanceGeek]

I've been spending some time studying various "lazy portfolios" and one of the ones that seems to test out quite well is Paul Merriman's ETF choices from FundAdvice.com - Suggested Portfolios

I took his recommended balanced ETF portfolio and rebalanced it to 70/30 (versus his balancing of 60/40) to get:

7% each in SPY, VTV, IWC, VBR, VNQ, EFA, EFV, and VWO
14% in DLS
15% in IEF
9% in SHY
6% in TIP

Then I started to think of ways to optimize the individual holdings within each sector. A few ideas there:

1. I don't think I can improve much on the Vipers he chooses, they have a possible tax disadvantage in that they are different classes of VG's regular index funds, but I see that as only a theoretical disadvantage and in any event I feel it is far compensated for by the superior expense ratios sported by the Viper family. Seems like nobody beats VG on cost.

2. I would substitute IVV for SPY. It has a lower expense ratio than SPY, and IVV has an advantage in that they reinvest dividends received from portfolio companies rather than distributing them to the shareholders of the ETF.

3. Instead of purchasing EFA, I would buy VGK / VPL in a 74 / 26 mix (well OK I'd probably buy it 75 / 25). You obtain the same mix that way with lower expenses, at the expense of extra brokerage trade costs.

4. I looked hard at FDM / PZI in place of IWC. But these seem too "actively" managed to me, and in any event IWC has much more liquidity and a lower expense ratio. I'll leave IWC in place.

5. Looked at RWR in place of VNQ, but again VG's lower cost structure wins out for me. They track different REIT indices anyway.

6. Sadly there are no "Government Bond" VIPERS to replace IEF/SHY/TIP with, but each is only 15 bps expenses anyway, so not much to save.

7. The position I'm least comfortable with in the resulting portfolio is DLS. Relatively high expense ratio of 58 bps and its only been out about 14 months. Since Wisdom Tree works off home grown and quasi actively managed indices its not going to be possible to find an exact match. Odd that a "dividend" fund only has a 0.36% yield according to Yahoo finance, but I suspect that the hassle of claiming foreign dividend tax credits would make it painful enough that I'd be happy this fund has a lowish yield. GWX might be an alternative, but its also expensive @ 60 bps and is very new.

Any thoughts to further improve/refine this portfolio?

 

[LOL!]

Here's what I think: You can try for lowest expense ratio and what not, but there are other things going on. You will need to occasionally do some tax-loss harvesting and you can buy intraday. How you do those will probably swamp any expense ratio stuff that you are looking at.

You may need to own EEM and VWO; GWX and DLS; IVV and SPY; RWR and VNQ because you may be doing tax-loss harvesting of some lots and replacing with a similar ETF.

Also look at the liquidity of some of those things. Once you have a 7-figure portfolio, you will own an entire days trading volume of the microcap IWC and 10% of the trading volume of DLS. These things can go several minutes without a trade or even an hour. So use limit orders to purchase. I myself own a six-figure slug of DLS and decided to not even bother with microcap because I feel it ain't worth it.

Anyways, don't get lost in details that will haver very little material effect in the end.

 

[FinanceGeek]

Thanks for the reply. I do think the expenses matter though. Going with VGK / VPL instead of EFA saves me 17 bps or so in expenses but adds another commission. If my round trip commission is $20, the break even point assuming a holding period of 1 year is a position size of $11700, not unreasonable. For a $100,000 position, that 17 bps difference pays for 1 trade a month, not the goal in this portfolio, but still significant.

Your point on liquidity is excellent, thanks. Limit orders certainly!

I like your idea of utilizing the various alternatives to take losses while avoiding the wash sale rule.

 

[LOL!]

You should not be paying any commissions at all for your ETFs. There are now ads all over this site for WellsFargo which gives you 100 free trades and a totally free checking account. If you are going to quibble about the difference in expense ratio of IVV and SPY (0.01%!!!) and then turn around and pay $10 per transaction, then you got some explaining to do.

Also, you do not want the dividends automatically reinvested. You want them in cash so that you can decide where to put the money (i.e. rebalancing!).

 

[justin]

6. Sadly there are no "Government Bond" VIPERS to replace IEF/SHY/TIP with, but each is only 15 bps expenses anyway, so not much to save.


Any thoughts to further improve/refine this portfolio?

It sounds like you might be aware of VG's 4 bond vipers. BSV - short; BIV - intermediate (7 yr avg maturity); BND - total bond mkt; BLV - long (20 yr avg maturity).

Expenses are around .11%. Not strictly govt bond funds, but all hold 50-70% govt/agency/govt backed bonds (some govt mortgage-backed).

These funds could end up being good substitutes for IEF/SHY. No VG equivalent for TIP's though. Taking a quick peak at the VG funds, the volume looked light today, so liquidity could be an issue (if that's an issue!).

 

[LOL!]

I forget to mention that there are other expenses with ETFs that everyone should be aware. First, the ETF can trade at a premium or discount to the NAV. Second, there is the bid/ask spread when transacting shares. I think that either or both of these can swamp expense ratios.

 

[FinanceGeek]

I forget to mention that there are other expenses with ETFs that everyone should be aware. First, the ETF can trade at a premium or discount to the NAV. Second, there is the bid/ask spread when transacting shares. I think that either or both of these can swamp expense ratios.

Good point, although with ETFs you don't have a huge disparity it certainly can be there and is significant for someone who's already counting basis points in expense ratios. Another solid reason to use limit orders.

On your other post about zero cost commissions I've always wondered what the true benefit of that really is - obviously the brokerage can make some of that up via lower rates on cash balances. And you still have to pay the SEC fees on sales right? In any event with large portfolios commissions for a lightly traded portfolio should be insignificant compared to spreads and interest on cash balances so I've always tended to look for quality of service overall. I've been with Fido and VG for years, and am happy with both.

 

[FinanceGeek]

It sounds like you might be aware of VG's 4 bond vipers. BSV - short; BIV - intermediate (7 yr avg maturity); BND - total bond mkt; BLV - long (20 yr avg maturity).

Expenses are around .11%. Not strictly govt bond funds, but all hold 50-70% govt/agency/govt backed bonds (some govt mortgage-backed).

These funds could end up being good substitutes for IEF/SHY. No VG equivalent for TIP's though. Taking a quick peak at the VG funds, the volume looked light today, so liquidity could be an issue (if that's an issue!).

Yep I looked at the bond vipers. Merriman's philosophy in creating this portfolio was to limit company risk factors to just the equity side of the portfolio, hence an insistence that the bond side be free of any default risk. He's a purist, I think one could take on minuscule risk by going for the vipers and pick up some incremental return in the process (right now BND has a 20 bps yield advantage over IEF). But it does muddy the waters a bit, and I understand why Merriman chose what he did.

On another point given our large deficits I sometimes wonder if the default risk of US bonds isn't under-appreciated, I suppose they'll just inflate it away, hence having TIPS in the portfolio provides partial immunization.

 

[LOL!]

I've been with Fido and VG for years, and am happy with both.

I've had mutual fund accounts at VG for more than 25 years. I hope you are not writing that are happy with their brokerage service which always rates poorly for good reason.

 

[clifp]

I also have IVV rather than SPY because the expenses are lower. One bit of warning while the liquidity of IVV is fine. SPY has active options, while options on IVV are very thinly traded. So if you want to write calls or hedge a portion of your portfolio with puts, you are better off using SPY.

The other Vanguard ETF to consider is VEU, it includes all international markets excluding the US. It has an ER of .15% and can be used to complement or replace VGK, VPL, and VWO.

 

[FinanceGeek]

I've had mutual fund accounts at VG for more than 25 years. I hope you are not writing that are happy with their brokerage service which always rates poorly for good reason.

Nope, I use VG for funds, Fido for brokerage and the odd (shhhhh!) active fund. Both perform their respective functions just great. VG's brokerage seems like an "also ran" to me, wonder if the move towards ETFs industry wide might convince them to improve its competitiveness a bit, e.g. that a substantial portion of their customer base might just decide to move to ETFs.

 

[FinanceGeek]

I also have IVV rather than SPY because the expenses are lower. One bit of warning while the liquidity of IVV is fine. SPY has active options, while options on IVV are very thinly traded. So if you want to write calls or hedge a portion of your portfolio with puts, you are better off using IVV.

The other Vanguard ETF to consider is VEU, it includes all international markets excluding the US. It has an ER of .15% and can be used to complement or replace VGK, VPL, and VWO.

I'm thinking you meant I'm better off with SPY for those reasons.

 

[justin]

Yep I looked at the bond vipers. Merriman's philosophy in creating this portfolio was to limit company risk factors to just the equity side of the portfolio, hence an insistence that the bond side be free of any default risk. He's a purist, I think one could take on minuscule risk by going for the vipers and pick up some incremental return in the process (right now BND has a 20 bps yield advantage over IEF). But it does muddy the waters a bit, and I understand why Merriman chose what he did.

On another point given our large deficits I sometimes wonder if the default risk of US bonds isn't under-appreciated, I suppose they'll just inflate it away, hence having TIPS in the portfolio provides partial immunization.

I'd put the credit risk of the VIPER bond funds at "very low". Especially the short term bond fund with its small exposure to corporate issues (30% of their portfolio) and the high credit rating of the corp issues they do hold (75% Aaa, 8% Aa, 10% A, 7% Baa). But you don't get a lot of extra yield for taking on a little extra risk.