Thursday 28 October 2010

Bond ETF Comparison: Canadian Broad Market Funds

Bonds are popular these days so let's have a look at three ETFs containing a broad range of corporate and government holdings. Bond funds are typically included in a portfolio to provide diversification for equities and other asset classes, to preserve capital by virtue of being more stable and secure than equities, or to generate income for spending from the interest they pay out on a regular basis. There are also many narrowly focused bond funds, such as high yield & risk, international bonds, short or long term maturities, corporate or government holdings, but broad ETFs provide a basic starting point and core holding for a portfolio.

The ETFs
Key Comparison Factors:
(for details click on the table image below)


Management Expense Ratio - ZAG is lowest overall by a hair at 0.28% over XBB. CAB's ratio is double that figure but its unique features may be worth it for certain investors with specific needs, as explained below.

Net Assets and Trading Costs - The winner here is XBB whose sheer size ensures a very active market for its shares. That keeps the spread between bid and ask prices very tight, which in turn minimizes investor trading costs (i.e. you can buy - the ask price - at a slightly lower price and sell - the bid - at a slightly higher price).

Diversification and Credit Risk - It's hard to conclude which is best on this dimension. On the one hand, XBB dominates on the number of holdings, which means the impact on the ETF of trouble at any single bond issuer is lessened. And it does have a higher percentage of its holdings in possibly safer government bonds than CAB. However, ZAG with the fewest number of holdings has the most in government bonds. CAB's strong suit is that, unlike ZAG and XBB, it holds none of the BBB or lower credit quality bonds that are most likely to run into trouble. Take your pick!

Interest Rate Risk aka Duration - Interest rates are as low as they have ever been right now but sooner or later they are likely to rise. Duration measures the sensitivity of the portfolio's bonds to changes in interest rates - the lower the duration, the lower the sensitivity (see Shakespeare's brief explanation here and a detailed mathematical explanation here on gummy's stuff of the Financial Webring). CAB is the least exposed with its lowest duration, but not by very much.

When interest rates do begin to rise, that will put downward pressure on the prices of bonds within the ETFs and the share price of the ETFs will fall as well. A mini preview of this effect occurred in April of this year when interest rates briefly spiked up about 0.5% only to fall back down 1% by today. The Google Finance chart below shows the price dip of our three ETFs followed by the upswing.


Yield to Maturity vs Distribution / Coupon Yield - Some investors mislead themselves by looking at the cash distributions (termed either the Distribution Yield or the Weighted Average Coupon percentage) paid out by the ETF as the investment return they will get. But that neglects the fact that the return of a bond or a collection of bonds is made up of the interest coupon (the cash part) and the difference between the price paid for the bond and the eventual payback of the principal of the bond upon its maturity (again, see the above Shakespeare link for a good explanation). The real economic return, which takes account of both factors, the one that we investors need to look at, is the Yield to Maturity. Note from our comparison table that the yield to maturity is much lower than the cash yield from the payout of interest coupons - about 2% less for ZAG and XBB and about 0.5% less for CAB. Don't be fooled - the 2.7% or so that the three ETFs give off is the real return you will be getting at today's ETF price and current interest rates. The low yield reflects the low interest rates in the economy.

There is very little difference between the three ETFs in yield to maturity, only 0.06% between the lowest ZAG and the highest CAB. That's not enough to call a winner.

After-Tax Cash Flow - For investors who buy bonds to receive cash income to spend, the cash distribution amount is a factor to examine. Here is where there is significant difference between CAB and the other two ETFs. ZAG and XBB simply hold bonds and pay out whatever coupon interest is received as they are obliged to do. ZAG and XBB have very similar payout rates of 4.6% and 4.7% respectively.

CAB has a unique structure, deliberately created by Claymore to fill a niche for a bond fund to be held in a taxable account. Most investors do not want or need that, having heeded the oft-repeated advice (e.g. at TaxTips.ca) to hold bond funds inside registered accounts where interest income does not get taxed until withdrawn. Some investors, perhaps having maxed out their RRSP, may want the stability and steady income of bonds outside a registered account. CAB uses a clever forward contract arrangement with TD Bank that allows it to distribute its income as a Return of Capital (ROC) for tax reporting. The cost of doing this - apparently TD receives a fee of 0.3 to 0.4% for its efforts - reduces the gross income but the after-tax net income is very competitive with the coupon yield of XBB and ZAG, especially for taxpayers in higher brackets e.g. the equivalent pre-tax distribution of CAB for a 40% tax rate is 5.34% versus only 4.7% for XBB.

That's because ROC is not taxed at all. Instead the ETF owner / taxpayer must reduce the Adjusted Cost Base of his/her ETF holding. The investor will eventually pay tax on the income as a capital gain instead of income at the time when he/she eventually sells the ETF holding (see our previous post on Calculating Capital Gains in ETFs and Mutual Funds). Claymore says it intends and feels it can achieve to only ever pay out ROC, and not income or capital gains from CAB. This is believable if the arrangement with TD or some other bank in its place is maintained indefinitely since then the returns (both income and price gains if any) will never get realized in tax terms. It's the same principle we outlined in our previous post Return of Capital: Separating the Good from the Bad as "good ROC" in the point about unrealized capital gains in mutual funds. At no point is CAB's income taxed at the higher marginal rate of interest - it achieves the minor miracle of turning the "water" of interest into the "wine" of deferred capital gains! The method used is totally legitimate and well established (the new TSX 60 index tracker from Horizons BetaPro which we reviewed here uses a very similar technique).

XBB and ZAG will produce interest income almost exclusively, though with a minor amount of good ROC - XBB gave off under 5% of its income as ROC in 2009 (see its Distribution History here). This is due to the growth of the ETF assets as investors buy in and more shares are created whose operation we described in our previous post on ROC under the heading Index Mimicking in ETFs.

Bottom line on this measure: XBB and ZAG are both fine and about equal for registered accounts while CAB is clearly best for a non-registered taxable account.

Automatic & Commission-Free Purchases and Withdrawals - For investors not wanting to receive and spend / withdraw the income, e.g. while in the portfolio buildup saving years, a DRIP that reinvests the income into the purchase of extra ETF shares automatically and without trading commissions is a big convenience and cost saver. ZAG and CAB both have the valuable DRIP feature while XBB does not. Two further valuable convenience features that only CAB offers are free and automatic pre-authorized chequing purchases of additional shares or systematic withdrawals by selling shares.

Which is best overall?
CAB is clearly superior for an investor who wants to hold bonds outside a registered account, especially those in higher tax brackets. Within registered accounts, ZAG and XBB are very similar whose slight differences offset each other. For someone building up a portfolio, ZAG is a hair ahead of XBB, due primarily to the DRIP.

Disclaimer: this post is my opinion only and should not be construed as investment advice. Readers should be aware that the above comparisons are not an investment recommendation. They rest on other sources, whose accuracy is not guaranteed and the article may not interpret such results correctly. Do your homework before making any decisions and consider consulting a professional advisor.

6 comments:

Anonymous said...

"XBB and ZAG are both fine and about equal for registered accounts while ZAG is clearly best for a non-registered taxable account."

Did you mean to say "CAB is clearly best for a non-registered taxable account"?

Anonymous said...

XBB is DRIP eligible through at least one financial institution -- iTrade.

CanadianInvestor said...

Yes, indeed CAB is the best ETF for a non-reg taxable account. I've now corrected the article. Thanks for the note.

jno said...

Thanks for this post. Two questions: If one is enrolled in a DRIP for XBB, does one still pay taxes on the interest income? And, if I have a choice to hold one of these funds in either a non-registered personal account or a business portfolio, which is better from a tax standpoint?

Many thanks!

CanadianInvestor said...

jno, yes the XBB interest would be taxed in a non-registered account even if re-invested through a DRIP ... though iShares does not offer a DRIP directly for XBB, it would have to be a broker doing it in a synthetic DRIP - see this post http://canadianfinancialdiy.blogspot.com/2008/01/driping-etfs-in-canada.html.

I'm not sure what you mean by a Business portfolio? Do you mean having XBB owned by a business that you own? If so, it is likely time to speak to your accountant to get some expert advice on what you could do, such as offsetting the interest against business losses, or if the business tax rate would be lower.

sleepydoc said...

It has been a while since you have looked at bond ETFs, and the 2013 budget essentially killed CAB. I would be interested in look into the options for holding bond ETFs in a taxable account. There have been new entries to address this issue such as BXF and ZDB. It would be great to hear your opinion!