ZDB vs ZAG 2026: The Tax-Efficient Bond ETF Choice for Canadian Non-Registered Accounts
Most Canadian investors who own a bond ETF in a non-registered account are quietly giving up 60 to 90 basis points of after-tax yield every year, and the fix is almost mechanical: swap the standard aggregate bond fund for a discount bond fund that holds the same investment-grade universe but with lower coupons. This post compares BMO ZDB and BMO ZAG, works the after-tax math at three marginal rates, and lays out when the swap is worth doing and when it is not.
The problem bond ETFs create in a taxable account
A standard Canadian bond ETF holds a mix of government and investment-grade corporate bonds and pays out their coupon interest every month. For a bond fund held in a non-registered account, those monthly distributions are reported on a T3 as interest income, which the CRA taxes at your full marginal rate. There is no dividend tax credit, no 50% capital gains inclusion, no reprieve of any kind. On $100,000 held in ZAG at a 4% yield-to-maturity, the pre-tax annual income is roughly $4,000, and at a 50% combined marginal rate the after-tax income shrinks to $2,000. That is a 2% after-tax yield on a fund the fact sheet describes as a 4% product.
The fix is not to abandon bonds, lengthen duration, or chase yield. It is to hold bonds that generate the same total return but with a different tax composition: lower coupon income, higher capital-gain accrual as the bonds pull toward par at maturity. That is exactly what the BMO Discount Bond Index ETF (ZDB) is engineered to do.
What ZDB and ZAG actually hold
ZAG (BMO Aggregate Bond Index ETF) tracks the FTSE Canada Universe Bond Index. It holds roughly 1,500 individual Canadian investment-grade bonds across federal, provincial, municipal, and corporate issuers, weighted by market value. Because interest rates have been higher than the coupons on older bonds for most of the last three years, a meaningful share of ZAG's holdings now trade above par (premium bonds) with coupons above the current market yield.
ZDB (BMO Discount Bond Index ETF) is a sibling fund that holds a filtered subset of the same investment-grade Canadian bond universe, selecting only bonds trading at or below par. The result is a fund with roughly the same duration, credit quality, and yield-to-maturity as ZAG, but with coupons that sit noticeably below the fund's YTM. The gap between coupon and YTM shows up as capital-gain accrual as the bonds mature at par, not as monthly interest.
Side-by-side specs (October 2026)
| Metric | ZDB | ZAG |
|---|---|---|
| Full name | BMO Discount Bond Index ETF | BMO Aggregate Bond Index ETF |
| Index tracked | FTSE Canada Universe (discount subset) | FTSE Canada Universe Bond Index |
| Management fee | 0.09% | 0.08% |
| MER (last annual) | 0.10% | 0.09% |
| Yield-to-maturity | approx 4.0% | approx 4.0% |
| Average coupon | approx 2.4% | approx 3.3% |
| Duration | approx 7.3 years | approx 7.3 years |
| Holdings | approx 440 bonds | approx 1,500 bonds |
| AUM | approx $1.3B | approx $9.0B |
| Distribution frequency | Monthly | Monthly |
| Distribution yield (last 12m) | approx 2.5% | approx 3.3% |
| Launch date | February 2014 | January 2010 |
| Bid-ask spread | approx 2 cents | approx 1 cent |
The after-tax math on $100,000
Assume both funds deliver a 4.0% total return over the year, split into coupon interest and price change. For a one-year snapshot, ZAG's return is roughly 3.3% interest + 0.7% price change, while ZDB's return is roughly 2.4% interest + 1.6% price change. Interest is taxed at full marginal rate; realized capital gains are taxed at a 50% inclusion rate (the 2024 federal budget's proposed 66.67% rate above $250,000 of gains was indefinitely deferred by the government in early 2025, so the 50% inclusion remains in effect for 2026).
| Marginal rate | ZAG after-tax yield | ZDB after-tax yield | ZDB advantage |
|---|---|---|---|
| 30% (middle bracket) | approx 2.90% | approx 3.12% | +22 bps |
| 40% (upper middle) | approx 2.52% | approx 2.84% | +32 bps |
| 50% (top combined bracket) | approx 2.13% | approx 2.56% | +43 bps |
| 53% (ON/BC top bracket) | approx 2.02% | approx 2.47% | +45 bps |
On a $500,000 fixed-income allocation held non-registered at a 50% marginal rate, that is roughly $2,150 of avoided tax a year, every year, with no change in duration, credit quality, or expected total return. Compounded over 20 years at a 4% YTM reinvested, the difference approaches $70,000. That is a mechanical structural edge, not a market call.
Account placement: ZDB only matters where it is taxed
| Account type | Hold ZDB or ZAG? |
|---|---|
| TFSA | Either, no tax difference. Default to ZAG for the slightly tighter spread and larger AUM. |
| RRSP / RRIF / LIRA | Either, no tax difference. ZAG often slightly better for the lower MER and tighter spread. |
| FHSA | Either, same reasoning as TFSA. |
| RESP | Either. RESP withdrawals are taxed in the beneficiary's hands, usually at a low rate, so the taxable-vs-sheltered distinction is largely moot during accumulation. |
| Non-registered (taxable) | ZDB, almost always. The 20 to 45 bps of after-tax yield is the entire point of the fund. |
| Corporate account (CCPC) | ZDB. Interest income inside a CCPC is taxed at the refundable rate (around 50.17% federally plus provincial), so the drag is at least as painful as personal top-bracket. |
Reasons to pick ZDB
- Non-registered account at any marginal rate above 30%
- You already max your TFSA, RRSP, FHSA and have taxable room
- Corporate investment account
- You want aggregate-bond exposure, not short-term or long-term tilt
- You can accept a 1-2 bp wider spread for 20 to 45 bps of tax savings
Reasons to stick with ZAG
- The position lives in a registered account
- You trade in and out often (ZAG's tighter spread matters more)
- You want the deeper AUM for institutional-size orders
- Your marginal rate is below 25% and the tax edge is small
- You already own ZAG with a large unrealized gain in taxable and the tax to switch exceeds a few years of savings
What about BXF, HBB, and the strip-bond ETFs?
BXF (CI 1-5 Year Laddered Government Strip Bond Index ETF) is often lumped in with ZDB as a tax-efficient bond fund. It is not quite the same thing. Strip bonds generate accrued interest every year that the CRA taxes as interest income on your T3, regardless of whether it was paid in cash. BXF's attractive distribution yield can mislead investors into thinking the fund is tax-advantaged, but the imputed interest shows up anyway. BXF is a reasonable short-duration bond holding, but it is not a tax trick.
HBB (Global X Canadian Select Universe Bond ETF) used to be the gold standard here because its total-return swap structure converted all distributions into capital gains inside the fund. CRA cracked down on that structure in 2022 and Horizons (now Global X) converted HBB to a physical fund in 2024, which eliminated the swap advantage. Today HBB is a normal bond ETF with no special tax treatment, so for taxable accounts in 2026 ZDB is the simplest and most durable option.
How to switch from ZAG to ZDB without triggering a tax bill
If you already hold ZAG in a non-registered account at a modest gain or loss, there are three clean ways to make the swap. The first is simple: sell ZAG and buy ZDB on the same day. For most investors in a flat-to-mildly-up bond market, the gain or loss triggered is small, and the perpetual future tax savings dwarf a one-time realization.
The second is tax-loss harvesting season (October to December). If ZAG is sitting at an unrealized loss, selling it crystallizes a capital loss you can use against gains in the current year or carry back three years. Buy ZDB immediately as the replacement; the two funds are not identical property under the superficial loss rule because they track different filtered subsets of the FTSE Canada Universe, so no 30-day waiting period applies. The IRS does not have jurisdiction, so US wash-sale rules are irrelevant for Canadian-domiciled funds.
The third, if the ZAG gain is large (say, over $5,000 of unrealized gain in a top-bracket investor), is to redirect new contributions to ZDB while leaving the existing ZAG position in place. Over three to five years the mix drifts to mostly ZDB and the forced tax hit never happens. This is the right play for investors with a very low cost base on inherited or long-held fixed-income positions.
The quiet question: does the discount actually hold up?
A fair criticism is that ZDB's edge depends on the Canadian bond market continuously generating a supply of discount bonds for the fund to buy. If rates fall meaningfully and stay low for years, the pool of below-par bonds shrinks and ZDB's filter forces the fund into a smaller, potentially less representative subset. Historically ZDB has tracked ZAG's total return within about 5 bps a year since 2014, suggesting the construction is robust across rate regimes, but it is worth monitoring in a sustained low-rate environment.
The other quiet risk is dividend-equivalent slip. ZDB publishes an annual reinvested distribution of capital gains (phantom distribution) that increases your adjusted cost base. If you do not update your ACB after each year-end, your eventual capital gain on sale will be overstated and you will overpay tax. Our ETF tax efficiency guide covers how to track phantom distributions mechanically; most brokers now report them on your T3 and in the ACB they display, but the responsibility is yours.
The switch rule
- Your bond allocation is in a non-registered or corporate account
- Your marginal rate is above 30% (true for most working professionals)
- You want aggregate Canadian investment-grade exposure (not short-term, long-term, or high-yield)
- You have no outsized unrealized gain in an existing ZAG position (or you plan to use the new-money approach)
- If all four are true: ZDB. If any is false: stay with ZAG or address the mismatch first.
A note on performance measurement
Pre-tax, ZDB and ZAG should and do deliver near-identical total returns. Any one-year gap is normally driven by the slightly different maturity bucket mix inside the discount filter, not by skill. Over the five years ending September 2026, ZDB and ZAG are within approximately 5 bps a year of each other on a pre-tax basis. The 20 to 45 bps a year of after-tax advantage therefore shows up only on your T1, not on a Morningstar chart. If you are evaluating the two by looking at pre-tax price charts, you will miss the entire point of ZDB.
Frequently asked questions
Is ZDB really the same bonds as ZAG, just filtered?
Not exactly. ZDB holds a subset of the FTSE Canada Universe filtered to bonds trading at or below par, with construction rules that keep its duration, sector weights, and credit quality close to ZAG. The duration and credit profile are nearly identical, but ZDB holds roughly 440 bonds versus ZAG's 1,500, so it is less diversified at the issuer level while equally diversified at the sector and quality level.
If ZDB is strictly better in taxable accounts, why does anyone hold ZAG there?
Three reasons. First, inertia: investors bought ZAG before ZDB existed (ZDB launched 2014) and have never changed. Second, large unrealized gains: a long-time holder may not want to trigger the gain to switch. Third, misunderstanding: distribution yield headlines make ZDB look like a lower-yielding fund, so investors avoid it without working the after-tax math.
Does ZDB generate phantom distributions I need to track?
Yes, usually once a year in late December. These are reinvested capital-gain distributions that are not paid in cash but are still taxable, and they increase your adjusted cost base. BMO publishes the per-unit amount on their website and most Canadian brokers now apply it automatically to your ACB. Confirm by checking your December T3 and your broker's ACB display.
What is ZDB's management fee vs its MER, and why do they differ?
The management fee is 0.09% and the MER (last annual, 2025 financials) is 0.10%. The MER adds expenses other than the management fee, chiefly HST on the management fee itself and small trading costs. The gap is normal for passive ETFs. Compare MERs when ranking funds; the fee alone understates the all-in cost.
Can I use ZDB in a corporate investment account?
Yes, and it is often more valuable there than in a personal non-registered account. A CCPC pays passive-income tax at roughly 50.17% federally plus provincial on interest income (part of which is refundable), so cutting the interest share of your bond fund's distribution is a direct refundable-tax drag reduction. Pair ZDB with eligible-dividend-paying equity ETFs in a CCPC for best after-tax results.
Should I buy ZDB inside a wrapped balanced fund like VBAL or XBAL?
You cannot pick the bond component inside a wrapper ETF. VBAL and XBAL hold standard aggregate bond sub-funds (VAB and the iShares equivalent), not discount bond versions. If tax efficiency in a non-registered account is a priority, build the balanced portfolio from component ETFs (VEQT or XEQT plus ZDB) rather than using a wrapped balanced product.