So, if you caught Part 1 of our series, “Tax-Loss Selling With ETFs”, you now know the basics of how a market downturn can actually be a tax-savings opportunity in disguise.
But how often should you check your portfolio for these “opportunities”, and when does it make cost-effective sense to take them?
Many investors – and advisors – pay little attention to tax-loss selling until year-end. By doing so, they miss opportunities that arise from market downturns throughout the year.
Consider an investor who purchased $100,000 of the iShares Core MSCI Emerging Markets IMI Index ETF (XEC) in April 2013. XEC was significantly down in the third quarter of the year, but during the fourth quarter it made an impressive recovery. Our investor could have crystallized a loss of more than $6,000 in July or August, but by the end of the year the opportunity had vanished.
This is why we recommend keeping an eye on your non-registered accounts year-round, not just in December. Pay particular attention when you know your holdings are probably taking a hit. Market downturns aren’t usually a big secret.
Next, how big of a capital loss do you need to realize to make it worthwhile? As there are additional trading costs of this strategy, such as bid-ask spreads and commissions, it should be large enough that the ongoing tax benefit easily outweighs the one-time expenses from implementing the strategy. In his book The Only Guide You’ll Ever Need for the Right Financial Plan, Larry Swedroe suggests two requirements that should be met before engaging in tax-loss selling. The security should have:
1. A minimum dollar loss of $5,000, and
2. A minimum percentage loss of 5%
As an example, suppose you originally purchased $100,000 of the Vanguard FTSE Canada All Cap Index ETF (VCN) in your non-registered account and your holding is currently showing a dollar loss of $5,000 (calculated as Market Value minus Book Value). This loss meets the minimum dollar threshold of $5,000. VCN also has a percentage loss of 5% (which is the $5,000 loss divided by the $100,000 book value). This loss also meets the minimum percentage threshold of 5%, so both rules have been satisfied. To realize the loss, you could sell all your shares of VCN and immediately purchase an equivalent amount of a similar, but not identical security.
| Swedroe’s tax-loss selling rule of thumb | Tax-loss selling threshold |
|---|---|
| Minimum dollar loss on the security, and | $5,000 |
| Minimum percentage loss on the security | 5% |
Bid-Ask Spreads Can Add Up
We prefer Swedroe’s double-barrel rule better than a single shot at specifying only a threshold dollar amount of loss. Also requiring a percentage threshold ensures the bid-ask spread cost of trading ETFs won’t offset the potential tax advantage.
Without a rule specifying a minimum percentage loss on a security, a $5,000 loss on a $100,000 holding would be treated the same as a $5,000 loss on a $1 million holding. In both cases, the tax deferral benefit is the same, but the bid-ask spread costs on the larger holding will be 10 times higher. So, if the bid-ask spread cost of a $100,000 tax-loss selling trade is $160, its cost would increase to $1,600 for a $1 million tax-loss selling trade. A trading cost of this magnitude would certainly offset the tax benefit for many years.
Market Recovery During 30-Day Period: Are You Feeling Lucky?
We currently use Swedroe’s $5,000/5% tax-loss selling thresholds with our PWL Toronto clients, but with one minor tweak. If we’re planning to switch back to the original ETF after 30 days, we increase the minimum dollar and percentage loss thresholds to $10,000 and 10% respectively. This adjustment considers the impact a strong market recovery during the 30-day holding period could have on the results of our strategy.
| Bender’s adjusted tax-loss selling rule of thumb | If switching back to the original ETF | If continuing to hold the replacement ETF |
|---|---|---|
| Minimum dollar loss on the security, and | $10,000 | $5,000 |
| Minimum percentage loss on the security | 10% | 5% |
Market recoveries during the 30-day period can be bittersweet. It’s sweet that your portfolio value is up. But you’ll face a bitter capital gain when you switch back to your original ETF after 30 days. This can offset some, or potentially all the benefit from your initial tax-loss selling trade.
Imagine you sold $84,000 of VCN on Christmas Eve 2018 at a loss of $6,000 and replaced it with XIC. In January 2019, the Canadian stock markets experienced an impressive recovery. If you had sold your replacement shares of XIC after 30 days, they would have been worth $93,000, resulting in a capital gain of $9,000 (which more than offset your initial $6,000 capital loss). Had you instead increased your tax-loss selling thresholds to $10,000 and 10%, this situation could have been avoided.
A significant recovery in the markets is hardly the worst thing that could happen, but it would still defeat the purpose of tax-loss selling. This is another reason to consider not switching back to your original ETF after the 30-day period has passed. If this is not possible, consider increasing your tax-loss selling dollar and percentage thresholds to $10,000 and 10%, respectively.
When we introduce you to our tax-loss selling pairs later in this series, we’ll let you know which of the original ETFs you should probably switch back to after 30 days, and which replacement ETFs you could consider holding for the long-term.
Banking Capital Losses: Losses Saved Are Losses Earned
I’ve also heard advisors say that it only makes sense to realize capital losses if their clients have already realized capital gains in the current year, or in the three prior years. They see no benefit to banking capital losses for later use.
I disagree with this logic, for a number of reasons.
First, many Canadian-listed ETFs tend to distribute capital gains at year-end, and these gains can be significant. By anticipating the need for capital losses to offset gains from these transactions, you can leave more money in the portfolio to grow.
Second, most investors will need to periodically rebalance their portfolios by selling equities, which will usually trigger capital gains in the process. Having an arsenal of capital losses to offset these gains will also come in handy when filing your future tax returns.
Finally, having capital losses carrying forward provides you with more portfolio management options. As new and improved ETFs are released, you may want to switch from a more expensive, less tax-efficient ETF to a cheaper or more tax-efficient ETF at some point in the future. We experienced this with our PWL clients when some Canadian-listed international equity ETFs began holding individual stocks directly back in 2014, making them more tax-efficient. For those clients with capital losses “in the bank”, it was easier to switch to the better products with fewer capital gains implications.
Next Up: Two Takes on Tax-Loss Selling – Justweath vs. PWL
To put all these tips into context, I thought you might appreciate reading some additional perspectives on how to implement tax-loss selling. Good news! You’ll find those extra viewpoints in Part 3 of this series, up next.
Hi Justin,
For your points 2 and 3 on reasons to do Tax Loss Harvesting (and rebalancing if there are new ETFs), I’m not sure I understand how they would work in practice. For example, if I purchased 100k of ETF XYZ and it drops to 70k. I then tax-loss harvest and purchase ZYX, which we expect to track XYZ very closely.
We have three possible scenarios:
1) ZYX increases in value to 120k (50k appreciation). We sell and apply our tax loss harvest, bringing the gain down to 20k. We can expect that if we had held XYZ, it would have also increased to 120k, for an identical gain.
2) ZYX stays at the same 70k. We have no gain to offset and we keep our 30k loss to use in the future. If we still had XYZ, it would also be at 70k, so now we would sell and get the 30k credit. It is identical.
3) If ZYX falls even further, to 50k and we sell it, we get another 20k loss credit, to 50k total loss. That would be the same had we kept XYZ.
Am I missing something?
Hi would like to take a tax loss on DXP and replace with ZPR to stay invested in the asset class. Thoughts? Also will you publishing a List of Tax Loss Selling ETF Pairs for 2019 like you did in 2016?
@TIM D: DXP is an active preferred share ETF, so it would not be an ideal tax-loss selling replacement for the passively managed ZPR (in fact, preferred share ETFs do not currently have ideal tax-loss selling replacements that I would feel comfortable recommending).
My fourth instalment in this blog series (release date of December 23, 2019) will include an updated 2019 tax-loss selling ETF pair list (I also discuss my recommended pairs in episode 2 of the podcast, which is now available).
Mark, since it was a hypothetical you that racked up the losses, probably unbeknownst to the real you, you have nothing to worry about and your egregious stock picks would be a moot point of little or no consequence.
Your assertion is correct that the value of the loss will be eroded but rather due to the compounding effects of time and money. Assuming you are able to reduce your taxes payable this year rather than next means that you have more money available for investing sooner, compounded over years. This premise also assumes that the real you doesn’t realize a loss but then turn it into a bigger loss, which can happen.
In general, realizing a loss after 40 years would not be considered prudent money management. You need to speak with the future you and prevent this from happening.
Hey Justin, really great podcast! Certainly changed the way I think about capital losses.
I wanted to explore the idea of “carrying forward capital losses indefinitely”. Suppose (hypothetical of course) that a younger me racked up say $15,000 in capital losses in a non-registered account by making bad stock picks. Now, suppose that I am investing in a much smarter way, but only in a TFSA / RRSP. It seems like I would never be able to take advantage of those banked capital losses (unless perhaps I leverage a LoC into a taxable account, but that is another story).
Or alternatively, suppose I now invest in a non-registered account, but always buy/never sell (until retirement). So, that capital loss from the age of 25 might be carried forward for 40 years to the age of 65 before it is realized. My question is, as the years tick on, doesn’t inflation erode the value of that capital loss, and if so, how can that eroded value be quantified?
@Mark H: You can also use net capital losses in the year of death and the prior year to offset income from those years (so the $15,000 capital loss might not be totally lost):
https://www.canada.ca/en/revenue-agency/services/tax/individuals/life-events/what-when-someone-died/net-capital-losses/net-capital-losses-year-death.html
The value of the loss would be the present value of the tax deferral in the future (so yes, presumably it would be worth much less than today).
Thank you for the great posts! You said “preferred share ETFs do not currently have ideal tax-loss selling replacements that I would feel comfortable recommending”. Do you also mean that it isn’t worth it to do tax loss harvesting on prefs share etfs?
@Erin: Glad you like them!
It’s not that I would recommend against tax-loss selling with preferred shares – it’s just that since they don’t have a suitable pairs, investors are taking more tracking error risk over the 30 days (i.e. their replacement ETF may underperform their original ETF by a noticeable amount). Investors should consider this before adding preferred share ETFs to their portfolio (especially if they are also planning to implement a tax-loss selling strategy).
Hmm, so by extension shouldn’t *all* tax loss selling decisions take into consideration the present value of the tax deferral in the future?
@Mark: There’s no way to know all of these future details. I think it’s safe to say that tax-loss selling with ETFs is usually best-suited for an investor (in a high tax bracket) with a large taxable account who has recently realized significant gains in the past three years. Even if the investor carries forward some of the net capital losses, they can still use these to offset annual capital gains distributions from the ETFs.
If this doesn’t sound like you, and you’re still wondering whether tax-loss selling can work for your specific situation, it’s probably not that useful.