Commentary and analysis on matters related to beta including indexing, exchange traded funds (”ETFs”), derivatives, their application in the portfolio management process and their effect on the investment industry.

Vanguard Bond ETFs: No Surprises, Lowest Costs

The fixed income ETF space is another area where we could see some serious expansion. International bonds, even in emerging markets, has been of great interest to a wide variety of investors in the global search for yields. So the good news is that along with BGI’s release of new fixed income ETFs on top of the six they had at the end of 2006, Vanguard is coming along with their first foray into the space. Vanguard has filed a registration statement with the SEC to offer ETF versions of four existing Vanguard bond index funds. Vanguard ETFs are structured as separate share classes of an existing Vanguard mutual fund counterpart.

The bad news is that Vanguard isn’t really venturing into new territory. Here’s the lineup with a table from Vanguard’s website:

Vanguard table 1

With an expense ratio of 11bps, these are cheap especially in comparison with the appropriate iShare counterpart. For example, for the broadest exposure, Vanguard’s Total Bond Market ETF is nearly half the cost of AGG which sits at 20bps. They both track the Lehman Brothers Aggregate Bond Index.

Comparisons between the remaining three Vanguard ETFs with their iShare counterparts is not as simple. First, the Lehman 1-3 Year Treasury Bond Fund (SHY), Lehman 7-10 Year Treasury Bond Fund (IEF) and Lehman 20+ Year Treasury Bond Fund (TLT) all have a 15bps fee which is not that far off Vanguard’s proposed 11bps. Next, you’ll note the difference in benchmarks:

Vanguard table 2

In this environment, and at first glance when looking at this table, you might not expect to see any significant performance differences between the Vanguard and iShare ETFs within each of the three groupings. However, a little digging on their respective websites leads to some interesting findings. The first thing I notice is the number of holdings in each fund.

Vanguard table 3

I couldn’t find data as of the same date, but despite this you still see quite a difference here. Just as an example/proof, the page from the iShares website shows the 10 holdings in TLT and, yup, they’re T-bonds going out about 20 to 30 years.

TLT’s ETF counterpart from Vanguard will be a share class of their Long-Term Bond Index Fund [VBLTX]. Although their data is a bit dated (as of September 30th, 2006), it does confirm that the number of holdings is roughly around 650 securities and lists them

Unlike TLT which is limited to US treasuries, Vanguard’s version is quite diversified in terms of type of issuer, maturity and credit quality.

Vanguard table 4

Vanguard table 5

Vanguard table 6

A quick review of the other Vanguard mutual funds (VBISX, VBIIX, VBLTX) show a similar degree of diversification far beyond what iShares provides. So, although at the very beginning I said that Vanguard is not venturing into new territory, they are providing something different and at a low cost.

So, for the short-term fixed income group, here’s the 3-year comparison price chart:

short term fixed income 3 yr comparison

Here’s the 3-year comparison price chart for the intermediate-term fixed income group:

intermediate term fixed income 3 yr comparison

Here’s the 3-year comparison price chart for the long-term fixed income group:

long term fixed income 3 yr comparison

Not too much to comment on with these three charts. In the chart for the short-term funds, we see greater month-to-month volatility in Vanguard’s fund versus the iShare which is not surprising. What surprises me a bit is how similar the month-to-month volatility is for the 2nd and 3rd charts. There are a small number of months, like April 2005, where there’s a bigger jump in the iShare versus the Vanguard fund, but otherwise they move almost in lock step.

Last thought. I wonder why Vanguard didn’t enter with a real return bond ETF? Vanguard’s Inflation-Protected Securities Fund Investor Shares [VIPSX] tracks very closely with iShares’ Lehman TIPS Bond ETF (TIP):

VIPSX TIP comparison chart

With TIP priced at 20bps, Vanguard could do what they did against AGG with an 11bps fee (along with other situations like (EEM) versus (VWO)) and force BGI to consider their overall ETF pricing strategy. With many of the recent and proposed ETF offerings venturing into new spaces, and with their associated relative higher costs, I’d like to see BGI go toe-to-toe with Vanguard in the more traditional asset classes and bring costs down.

Comments (4) left to “Vanguard Bond ETFs: No Surprises, Lowest Costs”

  1. DanH wrote:

    Richard,

    Have been reading your comments on ETF Investor for some time now. V pragmatic and no hype.

    I have been looking for a way to compare the cost of owning a bond ETF vs buidling my own ladder. Have seen some references to this in the FT but no definitive study comparing returns after fees. I understand the ETF manager should be able to get a better yield spread than a retail investor but by how much - and does this justify the MER? Many variables to consider - specifically duration as it drives buy transactions in the ladder costing yield spread every time. Given the MER is annual, do you have any view on if an ETF provider can cover the MER with improved yield? Any studies to point to. Thanx.

  2. The Beta Brief wrote:

    […] The idea of an equity index as a highly diversified means for market exposure is straightforward. With bonds, it’s just not the same. Bond indices can consist of hundreds or likely even thousands of underlying bonds however the correlations among the bonds is quite high … significantly higher than among stocks within an index. Because the gross returns among bonds are not that great, there is greater importance to costs within a bond ETF. The Canadian bond ETFs had a 25 bps MER. Even back in 2000 when the ETF industry was still in its relative infancy, I had many discussions with BGI Canada suggesting that they drop their MER somewhere below 5 bps, which I thought was more appropriate than 25bps. They never did that. But the important thing to note is that with bond ETFs, costs matter. I think that’s the number one rule of thumb when considering fixed income ETFs. By the way, you’ll be glad to know that about four years later in December 2004, BGI Canada converted the 5-Year Government of Canada Bond ETF into one that tracks a short term (less than 5 year maturity) bond index. The 10-Year Government of Canada Bond ETF was converted to track a well accepted Canadian bond index that includes various underlying government and corporate bonds. Although I easily stand by comments about diversification within a bond ETF, anything beats an ETF with one underlying position! We’re at the point today where in Canada we have bond ETFs that cover various areas of the yield curve for government bonds as well as coverage for corporates and real return bonds, very similar to where the US iShares business was not too long ago before BGI introduced their second wave of fixed income ETFs. Incredibly, despite having raised well over $20 billion in assets in their six bond ETFs, BGI waited about four and a half years before launching this second wave. And during this time, no one came to market with competing products. I wrote back in January about Vanguard’s (still not yet to market) bond ETFs which provide lower costs but not too much more beyond that. You’ll note in that posting how I hope for expansion in the bond ETF space to push internationally but I end up being disappointed. With all the innovation in the ETF industry, what’s going on in the bond side? […]

  3. RBuck wrote:

    For an income portfolio, do closed end mutual funds deserve consideratuion ? For instance a portfolio consisting of equal weighted DVF, EVV, VVR and PPR. These were chosen for their Ulcer Performance Index from 2006 and having a maximum drawdown of less than 4%. Generally they show a dividend payout of 7 - 8%. Not perfect of course - needs a stop loss point on each holding.

  4. Richard Kang wrote:

    RB: Yes, that makes sense and I’ve mentioned in various forums (this blog, other sites and newsletters) that exposure to different asset classes that traditionally pay higher distributions (infrastructure is a good example) might be the way to go. Take a look at my most recent posting on sector rotation and I discuss this further. I don’t give specific names as I try to avoid too much of that on this blog but I think you’re on the right track.

Post a Comment

You must be logged in to post a comment.