Canadian Banks Now Exceed 25% of the TSX — The Concentration Almost Nobody Is Pricing

The share of the S&P/TSX Composite made up of Canadian banks has crossed above 25% — a level economist David Rosenberg described in mid-July as unprecedented, a near 2.5 standard deviation event, and beyond what fundamentals would support. That single statistic reframes what most Canadians think they own when they hold “the Canadian market.”

The dispersion is the story

The TSX is on pace to beat the S&P 500 for a second consecutive year. That headline has been widely reported. The composition underneath it has not.

Segment	2026 performance so far
Big Six banks	around +33%
TSX headline index	around +12%
TSX excluding banks	under +6%

On a year-over-year basis the Big Six are up close to 70%. Strip the banks out and the rest of the Canadian market has delivered a return that barely clears a GIC after inflation.

If you hold a broad TSX index fund, your 2026 return is overwhelmingly a bet on six financial institutions. That is not diversification. It is a sector position wearing an index label.

What you are paying for it

Valuation is where the argument gets sharper. On Rosenberg Research’s numbers, the group trades at roughly 15 times expected 2027 earnings against a historical average nearer 11, and at 2.7 times book against a ten-year average of 1.7 — close to a 60% premium to its own decade-long norm.

The fundamentals underneath are genuinely good. Second-quarter net income rose 25% at RBC, 34% at BMO, 23% at CIBC and 15% at TD. Return on equity came in at 16.4% at CIBC and 13.5% at BMO, the latter up from 9.8% a year earlier.

But look at where a meaningful share of that came from. Provisions for credit losses fell 36% year over year at RBC, 28% at BMO and 20% at TD. Releasing loan-loss provisions flatters earnings in exactly the periods when credit looks benign — and reverses in exactly the periods when it does not. That is the hinge the next section turns on.

The parallel to Korea is uncomfortable

On July 28, South Korea’s Kospi fell 10.84% in a single session because two semiconductor companies carry that index. Nobody had to sell Korea. They only had to sell Samsung and SK Hynix, and the index went with them.

Canada’s version is slower-moving but structurally identical. A market where one sector represents a quarter of the index and nearly all of the return has one dominant failure mode: the sector re-rates, and the index goes with it regardless of what the other 200-plus companies are doing.

Two specific risks on the radar

Bond yields and the mortgage renewal problem. Fixed mortgage rates in Canada track Government of Canada bond yields, which rose this month as markets priced the possibility of Bank of Canada hikes. Canadian banks navigated the 2025 renewal wave without the damage many forecast. Late 2026 and 2027 renewals face a different yield environment — and a book whose reported earnings currently benefit from provisions being released rather than built.

Momentum. The rally has become increasingly momentum-driven rather than fundamentals-driven, the same characterisation applied to US tech and Asian semiconductors before both corrected. Momentum is a real return factor. It is also the factor with the most abrupt reversals.

What a Canadian investor can actually do

None of this is a sell signal, and the Big Six remain among the best-capitalised banks in the developed world. But there are concrete diagnostics worth running:

1. Measure your true bank weight. Add your direct bank holdings to the bank weight embedded in every TSX index fund, dividend ETF and Canadian equity mutual fund you own. Canadian dividend ETFs are frequently 40% or more financials. The total is almost always higher than people expect — the portfolio tracker will add it up across accounts.
2. Check what your “diversifier” actually diversifies. Adding a second Canadian dividend fund to a TSX index fund mostly adds more banks. Run the two through the comparison tool and look at the overlap rather than the names.
3. Look at the ex-bank TSX on its own terms. Energy, materials and industrials have lagged badly in 2026. Whether that is an opportunity or a warning depends on your view of oil and Canadian industrial demand — but you should at least know you are underweight it.
4. Mind the account. Canadian dividends receive the dividend tax credit in non-registered accounts, which changes the after-tax ranking of bank shares versus US holdings materially. The dividend tracker will show what you are actually receiving, and where.

Frequently asked questions

Are Canadian banks overvalued?
Valuation is contested. What is measurable is that the sector’s weight in the index, its contribution to index returns, and its price-to-book premium to its own ten-year average are all at levels without clear precedent, which raises the cost of being wrong.

What is the Bank of Canada’s policy rate?
2.25%, held at the July 15 decision, with the summary of deliberations published July 29. Markets are nonetheless pricing some probability of hikes.

What is the mortgage renewal cliff?
A concentration of Canadian mortgages originated at very low rates coming up for renewal at materially higher rates. The 2025 wave was absorbed better than feared. The 2026–27 wave meets a higher yield curve.

How much of a TSX index fund is banks?
Above 25% of the S&P/TSX Composite as of July 2026, and typically higher again in Canadian dividend-focused ETFs, many of which run 40% or more in financials.

Bottom line

The TSX’s outperformance is real. Its breadth is not. Investors who believe they hold a diversified Canadian portfolio should verify that belief with a weights spreadsheet rather than a fund name.