Truist (TFC) Q3 2026 earnings review

Truist's deposits grow, but its loans shrink and buybacks slow

Truist's new chief executive sold the bank's riskiest car loans and took a loss on its bonds, while lending shrank. Profit per share excluding those one-offs rose 12% from a year ago on record wealth fees and flat costs, and Mike Lyons called the quarter solid. But revenue excluding one-offs slipped 0.1% from the previous quarter against a plan for about 1% growth, the first time in over a year the bank fell short of its own outlook.

At a glance
Profit per share, excluding one-offs$1.16 +12% from a year ago
Revenue, excluding one-offs$5.30 billion −0.1% from last quarter; the plan said about +1%
Average loans$325.5 billion −1.1% from last quarter
Fourth-quarter buyback plan~$1 billion $1.2 billion this quarter

⚖️ Verdict: 🔴 Bearish

The story got worse because capital meant for buybacks went into bond losses. The car-loan sale freed capital, Truist spent it on a $1.4 billion bond loss, and it now plans about $1 billion of buybacks next quarter, less than in any quarter this year. The good news: deposits keep growing, the lending margin stopped slipping, and the sale took the weakest credit off the books.

The question now is whether Lyons is shrinking Truist into a more profitable bank or just a smaller one. Either the LightStream proceeds go to buybacks and better loans, or they fund another bond loss while commercial lending keeps falling. The fourth-quarter capital ratio and buyback figure will tell.

🐂 Bull Case

🟢🟢 new MARGIN

Deposits Replaced Borrowings and Lifted the Margin

Truist funds its loans mainly with customer deposits and fills the gap with borrowings from wholesale markets, which cost more. This quarter the deposits did more of the work. Average balances rose 1.4% from the previous quarter and 3.6% from a year ago, led by interest-bearing checking accounts, the company said.

At the same time the bank repaid short-term borrowings, partly with proceeds from the car-loan sale, so their average balance fell 36% from the prior quarter. That lowered the cost of its borrowed money. So the net interest margin, the gap between what the bank earns on assets and pays for funding, rose to 3.00% from 2.98%.

One caution: deposits that pay no interest keep shrinking as a share of the total, to 25.1% from 25.6% last quarter. Truist lists an "unfavorable deposit mix" among the drags on the margin from a year ago, so the gain depends on checking balances staying put.

What to watch: the fourth-quarter net interest margin, due in January. A reading at or above 3.00% would show deposit growth replacing the income the sold car loans used to earn, rather than the margin simply pausing before another decline.

🟢 strengthening GROWTH

Wealth Fees Set a Record While Dealmaking Flattened

Truist earns fees from managing client money, arranging deals and trading, card payments and mortgages. The presentation lists a "Record quarter in wealth management income": that line reached $407 million, up 8.8% from a year ago, which the company credits to "favorable market performance".

The other fee lines did less. Investment banking and trading income matched a year ago and fell 8% from the previous quarter on lower trading. So fee income excluding one-offs grew 2.5% from a year ago, after double-digit growth in the first half. The comparison explains most of that, because the dealmaking unit had a strong quarter a year ago.

What to watch: fourth-quarter fee income against the roughly $1.63 billion the revenue outlook leaves for it by our math, about 5% above a year ago. Falling short would mean the wealth record rested on the stock market, since client assets are still below a year ago after a 2025 divestiture.

🟢 strengthening MARGIN

Flat Costs Kept Returns Above the Year's Target

Truist told investors in July to expect costs to rise about 2% this quarter. Excluding $40 million of car-loan sale costs, expenses rose just 0.3% from the previous quarter. The presentation calls that growth "well controlled", and the numbers support it.

Pay was the line that moved: personnel expense rose 6% from a year ago on higher benefit costs and incentive pay in revenue-producing units, the company said. Lower other expense, after a cut to accruals for operating losses, offset much of it.

Because revenue excluding one-offs held and costs did not rise, return on tangible common equity excluding one-offs reached 14.6%, above the 14% the bank set for 2026. Chief executive Mike Lyons said he remains confident in "our near- and long-term ROTCE targets".

What to watch: fourth-quarter costs, which Truist says will rise about 2.5%. A rise above that would mean the new chief executive is spending to "add to existing talent", as the presentation promises, before the revenue from that talent shows up.

🐻 Bear Case

🔴🔴 fading GROWTH

Commercial Loans Fell Before Any Sale Closed

Truist describes its loan trends as "optimization of lending portfolios". It stopped making boat and RV loans, is running off low-return books and sold its subprime car lender. Those choices explain only part of the decline. Average loans fell 1.1% from the previous quarter, to $325.5 billion.

  • Car-loan sale and run-off: $4.4 billion of the quarter's drop, by the company's own roll-forward
  • Underlying loan growth: $0.8 billion, or 0.2% by our math
  • Commercial and industrial loans: down 1.7% from the previous quarter, in a book the bank is keeping
  • A year earlier: average loans grew 2.5% in a single quarter

This matters because commercial lending is the business Truist is keeping, and it shrank by $2.8 billion in a quarter with no sale attached. Net interest income rose 1.1% from the previous quarter, but the quarter had one more day, which by our math accounts for almost all of that. The release itself names "lower loan balances" as a drag.

What to watch: average commercial and industrial loans in the fourth quarter, with no sale to blur them. A second decline would mean the shrinking reflects borrower demand rather than strategy, and the return target gets harder to reach without a balance sheet to earn it on.

🔴🔴 new CAPITAL ALLOCATION contradicts narrative

Bond Losses Consumed the Car-Loan Sale's Capital

Chief executive Mike Lyons took two large actions this quarter: Truist sold Regional Acceptance, its subprime car-loan unit, and sold most of its low-yielding bonds at a loss. The company calls the pair "modestly accretive to earnings, ROTCE, and tangible book value per share".

  • Car-loan sale: $5.5 billion of loans sold for $5.2 billion; capital created, $990 million
  • Bond sale: $8.6 billion yielding 2.67%, sold at a $1.45 billion pre-tax loss, $1.1 billion after tax
  • Bonds bought: $7.2 billion yielding 5.60%, a rate the presentation bases on the futures curve for the next 12 months
  • Income pickup: about $174 million a year by our math; the company puts it at $40 million next quarter
  • Capital ratio: 10.7% from 10.9%, as the bond loss outweighed the car-loan gain

The arithmetic is less kind than the label. The after-tax bond loss exceeds the capital the car-loan sale created, so together the two deals used capital rather than freeing it. By our math the income pickup takes about eight years to earn back the pre-tax loss. The bonds sold had a 5.8-year duration, so much of that loss would have reversed on its own as they matured.

The buyback paid for it. Truist now plans about $1 billion of repurchases next quarter, down from $1.2 billion in each of the last two. By our math the year then ends about half a billion dollars short of the target the bank set in July.

What to watch: whether Truist repeats the pattern after the LightStream sale, where the presentation says it will evaluate a bond repositioning "to offset capital created from sale". A second bond loss funded by a second loan sale would mean the capital these disposals free goes to the bond book, not to shareholders.

🔴🔴 persistent DISCLOSURE contradicts narrative

Credit Improved Because Truist Sold Its Worst Loans

The presentation says "Asset quality remains strong", and the headline numbers agree: nonperforming loans fell to 0.40% of the book from 0.51%, and the bank released loss reserves. Nearly all of that improvement comes from one event, the sale of Regional Acceptance, whose borrowers the company says carried more than a tenth of a point of the nonperforming ratio.

  • Charge-offs: 0.77% of loans annualised, or 0.38% excluding the sold unit, against 0.50% last quarter
  • Commercial and industrial loans no longer paying interest: $733 million, up 12% from the previous quarter
  • Commercial and industrial loans 30–89 days late: $229 million, up from $142 million
  • Loss reserve: 1.25% of loans, down from 1.51%, after a $581 million release tied to the sale

This matters because the sale removed a known problem while the remaining book shows a new one. The commercial loans Truist intends to keep are souring faster: nonaccruals and early delinquencies in commercial and industrial lending both rose in a quarter the bank reports as an improvement. The charge-off rate excluding the sold unit is low, but it measures losses already taken, not loans going bad now.

What to watch: commercial and industrial nonaccrual loans in the fourth quarter, and the detail on modified loans in the quarterly filing, which these materials do not include. Another rise would mean the credit cycle is turning in the part of the book that no sale can fix.

🔴 persistent GOVERNANCE

Risks this quarter didn't answer

Four standing concerns got no numeric reading in the release or the presentation this quarter, and a fifth got reassuring words without a date. Each needs a specific disclosure before it can move the case either way.

  • Return target timing: Lyons says he is confident in "our near- and long-term ROTCE targets", but the 16–18% long-term goal still has no date, and the presentation says "Broader strategic review ongoing".
  • Capital rules: no update on the proposal to count unrealised bond and hedge losses in regulatory capital; accumulated other comprehensive loss widened to $8.3 billion from $6.7 billion.
  • Modified loans in payment default: the commercial figure rose sharply over the past year; the update sits in the quarterly filing, not supplied here.
  • Hedge losses: cash-flow-hedge losses are due to move into earnings over the next year; the fourth-quarter outlook names only the car-loan and bond effects.
  • Goodwill: the wholesale unit's cushion was thin at the October 2025 impairment test; this year's result arrives with fourth-quarter results.

👓 Other Themes

new PRODUCT

Truist Is Selling Its Online Consumer Lender

Truist agreed on October 14 to sell $7.7 billion of unsecured consumer loans, nearly all of LightStream, its online lending platform, to Carlyle Group. The company expects $7.2 billion of proceeds and about $850 million of capital, and calls the unit a "loan only business generating a below target ROTCE". Closing is set for the fourth quarter, with proceeds repaying wholesale borrowings.

💲 Other KPIs

Diluted share count (26Q3) 1,226 million
⇘ decelerating

Down 5.4% from a year ago. The decline has widened for a year, from 3.9% in the third quarter of 2025, because the bank has bought back more stock than it earned after dividends. Next quarter's smaller buyback will slow the shrinkage.

Tangible book value per share (26Q3) $31.90
⇄ reversing

Fell 4.5% from last quarter and 2.1% from a year ago, the first year-on-year decline in the five quarters we hold. Tangible book value is equity minus goodwill and other intangibles. The drop coincided with a $1.6 billion widening in accumulated other comprehensive loss, the unrealised losses on bonds and hedges that count against equity.

Truist Assist chat volume (26Q3) 2.0 million
⇘ decelerating

Up 54% from a year ago but only 1.4% from the previous quarter, so growth is flattening. Truist Assist is the automated chat helper inside the bank's app. Digital transactions, such as transfers and bill payments, were 71% of all consumer transactions, up 2.9 points from a year ago.

🔮 Guidance

Q4 Revenue, taxable-equivalent, excluding one-offs ~$5.35 billion
🠇 cut from Full-year 2026 up 3.5–4% (July), which by our math left $5.42–5.52 billion for Q4
⇒ stable

Cut. The plan has revenue up about 1% from this quarter's $5.30 billion, and by a similar amount from a year ago. By our math that puts the full year near 3.2% growth, below the range the bank gave in July and did not restate. The gap is the sold car loans: they take $125 million off the quarter, and without that the figure would sit inside the old range.

Q4 Net interest income $3.65–3.7 billion
🠇 cut from Full-year 2026 up 1–1.5% (July), which by our math left $3.68–3.76 billion for Q4
⇒ stable

Cut. The middle of the range is $3.68 billion, almost unchanged from this quarter and below the fourth quarter of last year. By our math it leaves full-year growth near 0.9%, under the 1–1.5% the bank gave in July. The cut is about the sales, not the business. The sold car loans remove $125 million and the new bonds add $40 million. Without them the quarter would rise about 2.6% by our math.

Q4 Noninterest expense Up about 2.5% from Q3
🠆 unchanged from Full-year 2026 up about 1.75% (July)
⇗ accelerating

Unchanged in effect. Costs rise about 2.5% from this quarter's $3.07 billion excluding sale costs. By our math that lands the full year at the roughly 1.75% rise the bank promised in July. A year ago the quarter carried a $130 million legal charge, so against that base costs would fall about 1%.

Q4 Net charge-off ratio ~0.45%
🠆 unchanged from Full-year 2026 about 0.55% (July)
⇘ decelerating

New for the quarter; the full-year figure was not restated. The plan has losses at 0.45% of loans, down from 0.77% this quarter, because the sold car loans carried most of the charge-offs. By our math the full year lands near 0.58%, above the figure the bank gave, since this quarter's $317 million of car-loan charge-offs stay in the count.

Q4 Effective tax rate ~13.5% (~15.5% taxable-equivalent)
🠇 cut from Full-year 2026 about 14.5% (July)
⇄ reversing

Lower than the year's plan. The rate rises from this quarter's 10.7%, which the bond loss pushed down by shrinking taxable income, but stays under the 14.5% the lender expected for the year. Tax credits from renewable-energy investments keep it there, so profit after tax gets a lift that revenue did not provide.

Q4 Share repurchases ~$1 billion
🠇 cut from About $5 billion for full-year 2026 (July), which by our math left about $1.5 billion for Q4
⇘ decelerating

Cut. Buybacks slow to about $1 billion from $1.2 billion this quarter. The July target was not restated. By our math the year ends near $4.5 billion against the $5 billion target. The LightStream sale should add about 0.2 points to the capital ratio when it closes, so the open question is whether that capital goes to buybacks or to another bond loss.

Return on tangible common equity targets 14%+ for 2026, ~15% for 2027, 16–18% long term
🠆 unchanged from Same targets (July)
⇒ stable

Unchanged. Lyons said Truist remains confident in "our near- and long-term ROTCE targets". Excluding one-offs the return was 14.6% this quarter and for the year to date, so the 2026 figure is on track. The reported return was 7.2% after the bond loss. The long-term range still has no date attached.

Preferred stock dividends $42M in 4Q26, $112M in 1Q27, $43M in 2Q27, $112M in 3Q27
🠆 unchanged from $42M in 4Q26, $112M in 1Q27, $42M in 2Q27 (July)
⇒ stable

Unchanged. The cost of preferred stock alternates between about $42 million and $112 million a quarter, so profit to common shareholders dips in the first and third quarters of each year. Investors comparing quarters should adjust for it.

❓ Key Questions

What is the full-year revenue plan after the car-loan sale?

Truist gave a fourth-quarter figure but did not restate July's 3.5–4% full-year range. By our math the new quarter implies about 3.2%. Does the bank count the sold loans as a miss against that range, or as a reset of it?

Why are commercial borrowers falling behind more often?

Commercial and industrial nonaccruals rose $76 million and early delinquencies $87 million in a quarter the bank calls strong. Which industries drove it, and are modified loans in payment default still rising?

Will another bond loss follow the LightStream sale?

The presentation says Truist will evaluate a bond repositioning to offset the capital the sale creates. How large, at what loss, and over how many years would the income pickup earn it back?

When does the 16–18% return target arrive?

Lyons reaffirmed the targets but the long-term range still has no date. Does the ongoing strategic review include more business sales, and would those change the 2027 figure?

How much do hedge losses drag on fourth-quarter interest income?

Cash-flow-hedge losses are due to move into earnings over the next year. The fourth-quarter outlook names the car-loan and bond effects but not this one; is it inside the $3.65–3.7 billion range?