Do Winning Funds Keep Winning?

Summary

No. Funds that led their category from 2016 to 2020 did not keep leading from January 2021 to October 2026. They did worse than chance: of the 136 funds in the top quarter of their category in the first period, 48.5% fell to the bottom quarter in the second, and only 17.6% stayed in the top quarter. Picking funds by their past five-year record would have pointed a corporation in the wrong direction.

Two things did carry forward. Low fees: cheaper funds did better in the second period. And volatility: funds that were calm in the first period stayed calm in the second, 98 times out of 111. What did not carry forward was return per unit of risk, and calm funds generally gave up more return than they saved in volatility.

Against a low-cost TD index fund, only 4 of 453 active funds came out ahead in both periods, fewer than the 7 that pure chance would produce. Index funds were the most consistent funds in the whole sample: TD U.S. Index finished in the top half of its category in all 11 periods tested, and TD Canadian Index in 10 of 11.

The page closes with the funds this test supports for the three corporate investors of the profiles report, revised where the test changed the answer. To beat the market: CIBC Nasdaq Index, Fidelity Canadian Growth Company Class and Fidelity Global Innovators Class. To match it at low cost and tax: TD's Canadian, U.S. and International index funds. To fall the least: Fidelity Canadian Large Cap Class and Capital Group Canadian Focused Equity. Each mix is illustrative, built from past data, and not a recommendation for any particular investor.

Key Findings

How the Test Was Built

The test asks a simple question: if a fund was among the best in its category over one period, was it still among the best over the next one? The first period, used to pick the leaders, is the five calendar years 2016 to 2020. The second, used to check them, runs from January 2021 to October 23, 2026. Nothing from the second period was used to choose the funds.

The universe is much wider than the 19 families of the main report, so that the result does not depend on which funds were chosen. It started from all 6,875 Canadian-dollar share classes in ten Morningstar categories: Canadian, Canadian focused, Canadian dividend and income, U.S., North American, global and international equity, and Canadian, U.S. and global small and mid cap. It kept funds that are open to investors and were launched by the end of 2015, one series per fund (the series with the highest management fee, which is normally the advisor series), and removed currency-hedged, U.S. dollar and option-income versions. Trust and corporate class versions of the same portfolio were counted once. That leaves 544 funds from 77 fund companies, 25 of them index funds.

Each fund was ranked against the others in its own category, separately in each period. With four equal groups and no skill at all, a quarter of each group would land in each quarter of the second period, and half of the leaders would finish in the top half. The test is before tax, because distributions by type are not available before 2021; the fund tables further down add the after-tax results of the main report.

Result 1: Past Leaders Reversed

Where the funds of each 2016-2020 quarter finished from 2021 to October 2026 (top, second, third or bottom quarter of their category), as a share of the group. With no persistence, every cell would be about 25%.

Only 33.1% of the first-period leaders finished in the top half afterward, against 58.1% of the first-period laggards. The result is not a statistical accident: when the second-period rankings were shuffled at random within each category 2,000 times, not one shuffle left the leaders as badly placed as reality did. The same reversal appears within the advisor series alone and among active funds alone.

The cause is one year. The 2016-2020 leaders were mostly growth-style funds, and growth stocks fell hardest in 2022. A fund's 2016-2020 rank predicts its 2022 rank strongly in reverse (a rank correlation of -0.49), and with 2022 removed from the second period the link between the two periods is zero. So the reversal is a style turning, not skill turning into its opposite. But once the style is taken out, there is nothing left: no sign that good managers stayed good.

Result 2: Against a Low-Cost Index Fund

Each active fund was compared with the TD index fund for its market, or a mix of them (MER about 0.5% to 0.7%): TD Canadian Index for Canadian and Canadian dividend funds, 70% Canadian and 30% U.S. index for Canadian focused funds, half and half for North American funds, TD U.S. Index for U.S. funds, TD International Index for international funds, and 70% U.S. and 30% international index for global funds. Small and mid cap funds have no matching index fund in the sample and are left out of this comparison.

Index: the yearly return of the matching index fund or mix, 2016-2020 / 2021 to October 2026, in percent. "Beat 1st" and "Beat 2nd" count the active funds with a higher yearly return than the index in each period, before tax and after each fund's fees. If beating the index in one period said nothing about the next, about 7 funds would have beaten it in both by chance; 4 did. Of the 112 funds that were top quarter of their category in the first period, 3 beat the index afterward.

The comparison is not quite even, because the active funds' MER includes advisor compensation and the TD index series costs less. Charging the index 0.75 point a year more, roughly the same advice cost on both sides, 10 funds beat it in both periods, still fewer than the 16 expected by chance. Fees are the one thing that clearly persisted: across the whole universe, cheaper funds ranked better in the second period (a rank correlation of +0.21 between low MER and second-period rank).

Result 3: Risk Persists, Return per Unit of Risk Does Not

Volatility, measured as the spread of a fund's calendar-year returns, carried over strongly from one period to the next (a rank correlation of +0.44). Of the 111 funds in the least volatile quarter of their category in 2016-2020, 98 were less volatile than the index from 2021 to 2026. A fund that is built to be calm, through its holdings, its style or a cash or bond sleeve, tends to stay calm.

Return per unit of volatility did not carry over (a rank correlation of -0.22). Only 17 funds were in the top quarter of their category on this measure in both periods, half of the 34 that chance would give. And calm came at a price: only 5 of the 111 calm funds earned more than the index in the second period. In Canadian equity, for example, the calm quarter returned 12.24% a year with a spread of 10.7, against 15.85% and 13.0 for the index, slightly less return per unit of risk than the index itself. For an investor who wants to stay in equities and fall less, the calm can be bought fairly reliably; extra return on top of it cannot.

Fund Companies

The same test applied to fund companies: the share of each company's funds that were in the top half of their category in both periods (chance: about 25%), their average percentile rank from 2016 to October 2026 (50 is the middle, 100 the best) and their average MER. Companies with at least 15 funds in the universe.

RBC comes closest to company-level persistence: almost half of its 37 funds were in the top half in both periods, close to twice the chance rate. It also has one of the lowest average MERs. Fidelity's average rank over ten years is among the best, but its funds swung between periods, which fits its growth style. CI sits at the bottom on both measures. A company's record is a better starting point than a single fund's, because it rests on many funds and several managers, but it is still no guarantee.

The Active Funds That Passed

Four active funds beat the matching TD index fund or mix in both periods. A fifth, Fidelity Special Situations, was outside the universe because Morningstar lists it as limited (it closed to new investors on July 2, 2021); tested on its own, it was the top Canadian focused fund of 2016-2020 and ranked in the top 4% since. Yearly returns before tax; the index column shows the matching index fund or mix for each period. Fidelity Canadian Growth Company and Special Situations are Canadian focused funds, RBC Life Science and North Growth are U.S. funds, and the EAFE pool is an international fund. Every worst year came in 2022 except the EAFE pool's (2018).

Two cautions. First, a handful of passes out of 453 is what chance produces, so even these records are evidence of a manager worth watching, not proof of skill. RBC Private EAFE is a private-wealth F series with no advisory fee in its MER; with one added, it would barely have beaten the index in the first period. Second, the two Fidelity funds are run by the same lead manager, Mark Schmehl, who also runs Fidelity Global Innovators. The only active record that survived the test in Canadian focused equity belongs to one person, and his funds should be treated as one bet.

Profile 1: Beat the Market

The test says that a past winner is not a reason to expect future outperformance. So this profile is built from two sources of higher expected return that did hold up: a market exposure with a higher growth tilt, and the one active manager whose record passed. CIBC Nasdaq Index ranked in the top 6% of U.S. equity funds in both periods; it gets there by owning the largest U.S. technology companies, not by picking stocks. It made no taxable distribution in five years. Fidelity Canadian Growth Company passed the test outright, and its corporate class kept 16.58% a year after tax since 2016. Global Innovators, the global version of the same manager's process, started in 2018 and cannot be tested over 2016-2020.

Weights: half in CIBC Nasdaq Index and a quarter in each Fidelity fund, rebalanced each January. Yearly return after corporate tax on distributions, from January of 2021 or 2016 to October 23, 2026; drag is the yearly tax on distributions, 2021 to 2025 average; worst calendar year before tax, 2016 to 2025 (2022 for all four lines). Fidelity Global Innovators started in 2018: since January 2018 it kept 24.50% a year after tax, and the mix 20.53%. The Fidelity funds are the corporate class versions, Series B. The same columns are used for the other two profiles.

This replaces Fidelity Special Situations Class in the profiles report, which is closed to new investors. Half the mix goes to the index fund so that one manager carries half of it, not two thirds. That choice cost a little in the past: equal thirds kept 17.55% a year after tax since 2021 and lost 25.32% in 2022. The mix leans heavily on U.S. technology and lost about a quarter of its value in 2022; it suits only an investor who can hold through that.

Profile 2: Match the Market at the Lowest Cost and Tax

This is the profile the test supports most directly. Low cost was the one fund characteristic that persisted, index funds were the most consistent funds in the sample, and almost no active fund beat the matching index fund twice. TD U.S. Index was in the top half of U.S. equity funds in all 11 periods, and TD Canadian Index in 10 of 11. The profile is unchanged from the profiles report.

One third in each fund, rebalanced each January. The worst years came in 2018 for the Canadian index fund and in 2022 for the others. In the test, TD U.S. Index was in the top half of its category in 11 of 11 periods, TD Canadian Index in 10 and TD International Index in 8.

Equal weights hold a third in Canada and a third outside North America; a Canadian and U.S. pair alone kept more over this decade (15.02% since 2021 and 12.65% since 2016), because international markets lagged. The Canadian index fund's drag is an upper limit: Morningstar reports its payouts as unsplit income, taxed here at the full 50.17% rate, while most of it is Canadian dividends taxed at a refundable 38.33%.

Profile 3: All Equity, Smallest Falls

Volatility was the most persistent trait in the test, so this profile can be built with some confidence. Return per unit of risk was not, so the funds were chosen on the rare cases where both held. Only 17 funds were top quarter on return per unit of volatility in both periods; Fidelity Canadian Large Cap and Capital Group Canadian Focused Equity were two of them, and they were also among the least volatile funds of their category in both periods (Fidelity Canadian Large Cap in the calmest 1% each time). Capital Group ranked in the top 13% and then the top 3% of its category on return per unit of risk.

Fidelity Canadian Large Cap Class, Series B, and Capital Group Canadian Focused Equity (Canada), Series A, half in each, rebalanced each January; all three worst years came in 2018. The mix gained 6.1% in 2022, when the market mix of profile 2 lost 9.8%, and its worst calendar year since 2016 was -1.6%. The price was return in strong markets: 10.42% a year after tax since 2016, against 10.68% for the index mix of profile 2 and 12.65% for a Canadian and U.S. index pair.

This replaces two funds of the profiles report. TD North American Dividend and Fidelity Dividend Class were among the best on return per unit of risk in 2016-2020 but only average since, and TD North American Dividend became more volatile than most of its peers. Adding Fidelity Dividend Class as a third fund lowered every result of the mix (13.32% since 2021, 9.88% since 2016, worst year -2.1%).

The Three Mixes Against Simple ETF Investing

After tax in a Québec corporation, to October 23, 2026. The ETF lines are the same benchmark as in the profiles report: half in an S&P 500 ETF (VFV) and half in a TSX ETF (XIC), and the same split in two corporate class ETFs (HXS and HXT) that pay no distributions. The funds' MERs include advisor compensation; the ETFs are shown without advice and with a 1.15% advisory fee.

The worst years came in 2022, except for the third mix (2018). Each mix did its own job: the first kept the most, the third fell the least. The second trailed the ETF pair mainly because a third of it is international. The persistence test is the reason to read this table with care: these results were measured over the same years the funds were chosen from, and the test shows how often such results fail to repeat.

Correction to the Two Earlier Reports

While building this test, an error was found in the data of the main report and the profiles report. Morningstar keeps a duplicate record for some funds that reports their returns in U.S. dollars. For 53 of the 319 funds in those reports, that record had been used, so their returns were in U.S. dollars instead of Canadian dollars. Both reports have been corrected with the Canadian-dollar returns. The main conclusions hold. The largest change is Fidelity Canadian Large Cap Class: its return in 2022 was 13.6%, not 6.1%, and it tracks its trust version closely. The same error affected 85 funds of this test's universe and was corrected before the results above were calculated.

What This Test Does Not Show

The universe holds only funds that still exist. Funds that closed or merged between 2016 and 2026 are missing, and most of them were poor performers, so the real record of picking active funds is probably worse than shown here. The test uses calendar-year returns, so volatility rests on five yearly figures per period and misses falls within a year. Two periods are not many: a test that starts in a different year could find a different pattern, and the 2022 turn dominates this one. The ranking is before tax; after-tax results are available only for the funds of the main report and are shown in the profile tables. Suitability depends on a full review of the investor's situation, and the tax treatment of any strategy should be confirmed with the corporation's accountant.

Methodology and Data Sources

Data: Morningstar Canada, retrieved October 25, 2026: calendar-year returns for 2016 to 2025 and returns for 2026 to October 23 in Canadian dollars, management fee, MER, inception date and status for each share class. After-tax figures, tax drag and the ETF benchmark are those of the main report and the profiles report, after the correction described above.

• Universe: open Canadian-dollar funds in ten Morningstar equity categories with an inception date on or before December 31, 2015; one series per fund (highest management fee, ties to the oldest); currency-hedged, currency-neutral, U.S. dollar and option-income versions removed; trust and class versions of one portfolio counted once when every calendar return was within 1.5 points. 544 funds.
• Periods: 2016 to 2020, and January 2021 to October 23, 2026 (2026 counts as 0.81 of a year); annualized. Ranks are percentiles within each Morningstar category, computed within this universe.
• Random check: second-period ranks shuffled 2,000 times within each category, keeping the first-period ranks fixed.
• Index comparison: TD Canadian Index, TD U.S. Index and TD International Index, Series I, or fixed mixes of them rebalanced yearly, as described in Result 2.
• Volatility: standard deviation of the five calendar-year returns of each period. Return per unit of risk: annualized return divided by that figure.
• After tax: distributions from 2021 to October 2026 taxed at 2026 Québec corporate rates (interest and foreign income 50.17%, Canadian dividends 38.33% refundable Part IV tax, capital gains 25.09%, return of capital not taxed); years before 2021 use each fund's 2021 to 2025 average; no refunds credited. Capital Group Canadian Focused Equity was added for this report: its distributions, all reported by Morningstar as unsplit income, were taxed at 50.17%.
• Mixes: fixed weights, rebalanced each January, built from each fund's yearly after-tax return.
• All figures are illustrative only. Assumptions: no withdrawals, no new deposits, distributions reinvested, no advisory fee beyond the MER unless stated.

Past performance does not guarantee future results. Up-to-date performance data are available upon request.

Report dated October 25, 2026. Data source: Morningstar Canada.