TD's Record Quarter Hides the Felony Asset Cap Strangling Its Only Growth Engine

---

There is a particular kind of earnings beat that an experienced analyst should learn to mistrust: the one that arrives wrapped in superlatives from a company that, eighteen months earlier, became the first bank in the history of the United States to plead guilty to a felony conspiracy to launder money. On May 28, 2026, The Toronto-Dominion Bank delivered exactly such a quarter. Adjusted earnings of $4.2 billion, up 15% year-over-year. Adjusted diluted earnings per share of $2.38, up a remarkable 21%. Revenue of $16.04 billion. Record second-quarter earnings in Canadian Personal and Commercial Banking, all-time-high results in Wealth Management and Insurance and in Wholesale Banking, a fourth consecutive quarter of positive operating leverage, a 4% dividend increase to $1.12 per share, and roughly 19 million common shares retired under a $7 billion buyback. The press release reads like a redemption arc. The footnotes read like a hostage negotiation.

This article is not a claim that TD is going to fail. It is a claim that the market is being shown a growth story and sold a run-off story, and that the difference between those two narratives is where the risk lives. TD's headline EPS growth is real. But a meaningful slice of that 21% per-share figure is not the bank earning more money — it is the bank dividing a stalled pool of profit across fewer shares while its single largest engine of long-term growth sits under an indefinite, regulator-imposed cap that no buyback, dividend, or quarterly beat can lift. Below the surface, Canadian consumer credit is migrating the wrong way, the U.S. franchise is being actively shrunk by design, and the cost of staying out of prison runs through at least 2027.

The felony that froze the franchise

To value TD honestly you have to start where the bulls would rather not: October 2024, when TD Bank's U.S. subsidiaries pleaded guilty to felony charges including conspiracy to commit money laundering, and the bank agreed to pay approximately US$3.09 billion across regulators — roughly US$1.8 billion to the Department of Justice, about US$1.3 billion to FinCEN, and US$450 million to the OCC. American Banker's headline put it plainly: "Profits over compliance." TD's own internal failures allowed transnational criminal networks to move hundreds of millions of dollars through its branches, in some cases facilitated by employees, over a span of years. This was not a paperwork lapse. It was, in the government's framing, a business that treated compliance as a cost center to be starved while the U.S. retail engine was floored.

The fine, however damaging, is the part that heals. The part that does not heal is the structural punishment the OCC bolted on: an indefinite asset cap limiting TD's U.S. retail subsidiaries to roughly US$434 billion in total assets. This was unprecedented — the first time the OCC has imposed an asset cap as part of a BSA/AML consent order — and it comes with a loaded gun. The consent order permits the OCC, at its sole discretion, to compel TD to shrink its U.S. consolidated assets by up to an additional 7% every year the bank fails to satisfy its remediation obligations. Read that again. The regulator can make the franchise smaller, not just stop it from growing, if it judges the cleanup inadequate. There is no other large North American bank carrying that specific sword over its head today.

Where 21% EPS growth actually comes from

Here is the sleight of hand the headline obscures. TD grew adjusted earnings 15% but adjusted EPS 21% — a six-point wedge that exists because the share count is falling fast. The bank repurchased roughly 19 million shares this quarter alone under a $7 billion program. Buybacks are not inherently sinful; returning capital can be the single most rational thing a mature bank does. But the forensic question is always the same: what are you buying back the shares with, and why?

TD is returning capital aggressively in part because it cannot deploy that capital into its highest-return historical use — growing the U.S. retail bank — without bumping into a hard regulatory ceiling. Capital that would once have funded U.S. branch expansion, loan growth, and tuck-in acquisitions is instead being routed back to shareholders because the growth runway has been legally severed. The CET1 ratio fell 26 basis points sequentially to 14.3%, with the buyback alone consuming 41 basis points. That is a bank shrinking its own equity base to flatter per-share metrics precisely because the organic-growth alternative is foreclosed. A rising EPS line built on a falling share count against a capped asset base is the financial equivalent of a treadmill set to incline: the effort is real, the scenery never changes.

Selling fifty billion to fit inside the cage

The most telling disclosure is not in the earnings beat — it is in the balance-sheet repositioning. To comply with the asset cap and improve the yield on a constrained book, TD has committed to shrinking its U.S. assets by about 10% and to selling as much as $50 billion of lower-yielding investment securities, reinvesting the proceeds. This is being marketed as savvy optimization, and in fairness, swapping low-coupon legacy bonds for higher-yielding assets does lift net interest margin at the margin. But notice what it actually is: a forced contraction dressed as a strategy. You do not sell $50 billion of securities and shrink a franchise by a tenth because you found a brilliant opportunity. You do it because a federal regulator drew a line you are not allowed to cross, and the only way to keep earning inside that line is to reshuffle the deck rather than add cards.

The bank has separately moved to offload mortgage portfolios — reporting around the $9 billion mark — as part of the same strategic compression. Every one of these transactions improves a ratio in the short run and amputates a piece of the earning-asset base in the long run. A bank whose primary post-crisis "strategy" is figuring out how to be smaller and still look like it is winning is not a growth bank. It is a utility being managed for cash extraction while it serves a five-year probationary sentence with independent compliance monitors watching every move.

The remediation bill that does not show up in the beat

Pleading guilty bought TD a five-year probationary term, multi-year monitoring, an OCC-mandated U.S. compliance office, and a requirement to relocate the parts of its anti-money-laundering program responsible for U.S. compliance into the United States. Independent monitors now sit inside the bank. Management has signaled significant AML investment and remediation milestones stretching through 2027.

These costs are real, recurring, and structurally opposed to the operating-leverage story TD is selling. The bank touts a fourth straight quarter of positive operating leverage — revenue growing faster than expenses — yet it is simultaneously obligated to pour hundreds of millions of dollars into a compliance apparatus that produces no revenue whatsoever. Some of that spend lands in "adjusting items" and is conveniently excluded from the adjusted EPS the bulls quote. That is the quality-of-earnings problem in miniature: the number everyone cites as proof of a clean recovery is the number that has been scrubbed of the very expenses the felony created. When a company asks you to look at results "excluding" the cost of its own criminal remediation, you should at least insist on seeing both.

The Canadian engine, idling on consumer credit

With the U.S. growth runway capped, TD's narrative leans almost entirely on its Canadian retail bank — and that is where the second, quieter problem lives. Canadian Personal and Commercial Banking did post record second-quarter net income of $1,925 million, up 15%, on 5% revenue growth and 6% loan growth. Impressive on its face. But peer beneath it and the credit trend turns.

Total provisions for credit losses came in at roughly $1,001 million, about 43 basis points, within TD's own full-year guidance of 40 to 50 basis points. The composition is the concern. Impaired provisions — the ones tied to borrowers already in clear trouble — rose 9% to $465 million, which TD attributed to "credit migration" in its consumer portfolios. Performing provisions were driven by consumer lending and a darker macroeconomic outlook. This is a bank telling you, in its own footnotes, that its Canadian consumer is starting to crack at the exact moment it needs that consumer to carry the entire growth story alone.

The backdrop sharpens the worry. Canada's mortgage market faces a multi-year wall of renewals, with households that locked in at pandemic-era rates rolling over into materially higher payments. TD's own April 2026 survey found household budgets tightening amid those mortgage renewals. Real-estate-secured lending volumes grew 5% year-over-year — but volume growth into a renewal cliff, with impaired consumer provisions already rising, is precisely the setup where a "record" quarter today becomes a provisioning quarter tomorrow. The bank is leaning its whole weight on the one leg that is starting to wobble.

Cyclical relief, priced as secular redemption

Strip everything down and TD presents as a classic case of a cyclical recovery being capitalized as a secular turnaround. The bond-book repositioning lifts margin; the buyback lifts EPS; the legal settlement is in the rear-view; the dividend went up. Every one of those is a real, datable improvement. None of them changes the fundamental fact that the franchise's highest-return growth engine is legally frozen for the indefinite future, with a regulator empowered to make it smaller still.

Valuation is where this asymmetry bites. Depending on the lens, TD trades around a trailing price-to-earnings ratio near 18x — above a North American bank-industry average closer to 12x and a peer average near 16x — even as forward estimates compress that multiple toward the low-teens. In other words, on trailing earnings the market is paying a premium to peers for a bank whose defining feature is a unique, indefinite, regulator-imposed growth cap. A premium multiple is the market's way of pricing in superior future growth. It is difficult to think of a large bank with a more explicitly constrained growth future than the one operating under a felony plea, a five-year probation, an asset cap, and a clause letting its regulator shrink it 7% a year. Priced cheap on forward numbers, perhaps — but cheap can persist for years when the catalyst that would re-rate the stock is the lifting of a cap that has no announced expiration date.

The denominator illusion, in three acts

It is worth naming the recurring trick across all of this, because it appears three separate times in TD's story and each time it flatters the optics. Act one: EPS grows faster than earnings because the share count shrinks via buyback. Act two: net interest margin improves not because the bank is winning more business but because it is selling lower-yielding assets, lifting the average yield on a smaller base. Act three: ratios like return on equity get a tailwind as the equity base itself is deliberately reduced. In every act, the numerator is flat-to-soft and the denominator is doing the heavy lifting. None of the three is fraudulent. All three are the unmistakable signature of a franchise being optimized for appearance under a constraint, rather than expanded on its merits. When a company's best three growth metrics are all denominator stories, the growth is in the math, not the business.

What the bulls genuinely get right

The bear case has to be honest about its limits, because the bull case here is not a fantasy. TD is, by several measures, executing extremely well inside its cage. The Canadian retail bank delivered a genuine record — $1,925 million in net income on real loan and deposit growth, not accounting alchemy — and that franchise sits inside an oligopolistic Canadian banking system with structurally high returns and durable deposit moats that most U.S. regional banks would envy. The dividend increase to $1.12 signals real board-level confidence in capital adequacy, and a 14.3% CET1 ratio is a fortress by global standards, comfortably above regulatory minimums even after an aggressive buyback. The PCL of 43 basis points remains squarely within guidance; gross impaired loan formations actually decreased five basis points quarter-over-quarter, and gross impaired loans fell four basis points — so the consumer-credit deterioration, while real, is so far gradual rather than a cliff.

Crucially, the U.S. business is not collapsing under the cap; it grew adjusted net income to $960 million, up 8% year-over-year and 12% in U.S.-dollar terms — proof that TD can wring more profit out of a frozen balance sheet through pricing, mix, and efficiency. The asset cap forces discipline TD's pre-2024 growth-at-all-costs culture plainly lacked, and a leaner, higher-yielding U.S. book could prove more profitable per dollar than the bloated one it replaced. The bank also retains its valuable Schwab-related interests and a wealth franchise hitting all-time highs. And the single most powerful bull argument is simple optionality: the asset cap is indefinite, not eternal. If TD satisfies its remediation obligations — and it is spending heavily and visibly to do so — a future lifting of the cap would unleash a U.S. growth engine the market has written to zero. That is a genuine, asymmetric upside the bear case cannot dismiss. A patient owner is being paid a rising dividend to wait for an option that may well pay off.

Demonstration versus deployment

But optionality is not the same as deployment, and this is where the bull case quietly overreaches. The market is being invited to pay a premium today for a cap-removal that has no date, no guaranteed trigger, and a regulator with every incentive to move slowly. Remediation of a money-laundering failure severe enough to produce the first bank felony plea in U.S. history is not a checklist that clears in a clean fiscal year. It involves independent monitors, multi-year cultural overhaul, and a regulator that has explicitly reserved the right to shrink the bank further if it is unsatisfied. The base case for cap removal is measured in years, and the path is non-linear and entirely outside TD's unilateral control. You can demonstrate compliance every quarter and still not be granted your freedom on your own timeline.

That asymmetry — pay now, maybe get the growth back later, on the regulator's clock, with downside if the consumer cracks first — is the entire trade. The bulls are buying a call option and being charged the price of the underlying.

The moat that turned into a loophole

There is a deeper irony in TD's predicament that the redemption narrative glosses over entirely. For years, the bull thesis on TD rested on its U.S. retail franchise as a genuine competitive moat — a coast-to-coast deposit-gathering machine, "America's Most Convenient Bank," with long branch hours and a scale presence that smaller regionals could not match. That moat was supposed to be the growth differentiator versus TD's Canadian peers, most of whom never built a comparable U.S. footprint. The settlement reframed the entire story. What looked like a moat was, in part, a franchise that had outrun its own controls — scale that compliance never caught up to. The regulator did not just fine the bank; it identified the U.S. growth engine itself as the locus of the failure and capped the very thing that made TD special. The competitive advantage and the legal liability turned out to be the same asset.

That is why this is not a normal "buy the dip on a fined bank" situation. When a regulator fines a bank for a rogue trading desk or a mis-sold product, the franchise survives intact and the penalty is a one-time hit. Here, the punishment was surgically aimed at the growth engine, converting TD's chief differentiator into its chief constraint. The bull who buys TD for its U.S. scale is buying the precise attribute the OCC has frozen. The moat is still there on the map; the drawbridge is welded shut, and the keys are held by a regulator with a five-year memory and a clause to shrink the castle.

What a fairly priced version looks like

Imagine, briefly, the version of TD that the current trailing multiple would be reasonable for: a bank compounding its U.S. and Canadian books in the high single digits, redeploying capital into organic loan growth at attractive returns, with a clear runway and no overhang. That bank deserves a premium to the peer group. Now contrast it with the actual TD: U.S. assets shrinking about 10% by design, $50 billion of securities being sold to fit inside a cage, capital returned via buyback because it cannot be deployed for growth, Canadian consumer provisions migrating higher into a renewal cliff, and an AML remediation bill running through 2027. The gap between the premium the market is paying and the constrained reality is the margin of safety the bears are short of — and the bulls are long. Cheap on forward earnings is only cheap if the earnings are allowed to grow; under an indefinite cap, "cheap" can be a value trap that compounds slowly for years while the catalyst refuses to arrive on schedule.

The kicker

TD's quarter was good. That is precisely the problem: it was good enough to let everyone stop asking why a bank with a fortress balance sheet, a record retail print, and a rising dividend is also the only large lender in North America operating under a felony plea, a five-year probation, an indefinite asset cap, and a regulator who can legally shrink it 7% a year. The earnings beat is the anesthesia; the consent order is the surgery. And anesthesia, by design, makes you stop noticing the thing that is actually being done to you.

A bank can grow its per-share earnings every single quarter and still be quietly dying as a growth franchise, because the cruelest cap in finance is the one you are forbidden to grow past — and TD is the only one in North America legally wearing it.