Bank of Hawaii's Ten-Quarter Margin Machine Has a Mix-Shift Problem — and a New CEO Reframing the Story

Bank of Hawaii delivered the quarter it has been promising for two and a half years: net interest income of $156.1 million, a net interest margin of 2.82%, and a tenth consecutive quarter of expansion in both. Diluted EPS of $1.49 was up 24% year over year, return on common equity was 15.4%, and the efficiency ratio sits near 41%. And yet the shares are down 8.7% over the last 90 days and 17.8% from the July 17 peak of $85.85. That gap between the P&L and the tape is the story of this call — because the mechanical engine that produced those ten quarters is now mostly spent, and the inputs that used to help are quietly working against it.

Ten Quarters In, the Mix Flips

For years the Bank of Hawaii earnings model ran on two gears: fixed asset repricing — securities and loans maturing at low yields and rolling into higher ones — and a favorable deposit remix toward non-interest-bearing balances. The first gear still works. The second has reversed. CFO Brad Sattenberg was explicit: “This marked the 2nd consecutive quarter where the mix shift was negative, and although it accelerated this quarter, I expect it to moderate in the 4th quarter” — Bradley Satenberg, Chief Financial Officer · 2026-10-26. The four basis points of margin expansion came from asset repricing, "partially offset by an unfavorable deposit mix shift." Strip out one nonaccrual loan and NIM would have been 2.83%.

What replaces the mix tailwind is cyclical, not structural. The September rate hike gave only a sliver of benefit in Q3 but should add "several basis points" to earning-asset yields in Q4 as floating-rate assets reprice for a full period. That places Bank of Hawaii squarely inside the market's dominant macro motif — Fed rate hikes is the second-highest momentum theme across the entire editorial universe this quarter, echoed in the calls of ARR, SSB and a long list of regional banks. For a deposit-heavy lender, a hike is a near-term gift: liabilities reprice with a lag. The catch is that it is a gift with an expiry date, and it arrives just as the structural gear is slowing.

Note also what was not said. The terminal 3.25%–3.50% margin ambition that anchored prior calls — reachable "at the end of 2028" if fixed-asset repricing kept contributing 20 basis points a year — went unmentioned. The ten-quarter streak is real, but the framing has changed.

A New CEO Rewrites the Sentence

Jim Polk now runs the bank, and his prepared remarks contain the most consequential admission on the call. After praising the margin, he conceded:

At the same time, we recognize that sustaining earnings growth over the longer term will require a greater contribution from the organic deposit, loan, and fee income growth.

That is a different sentence than the one his predecessor used eighteen months ago, when Peter Ho described the repricing engine as “pretty mechanical at that 20 basis points a year, given a 10-year in the 4.25% range” — Unknown Speaker, President and CEO · 2026-04-20. Ho's version was a machine you could model. Polk's version is a franchise you have to sell. He spent most of his remarks on market share (34% of Hawaii deposits, up 6.7 points in twenty years), on wealth management — trust and asset management plus annuity and insurance fees of $45.2 million for nine months, up 11.5% — and on the Center for Family Businesses and Entrepreneurs. Effectively, the new CEO is telling investors that Hawaii's economy grows at low single digits (1.6% projected for 2027), so the bank must take share rather than ride the tide.

That reframing matters because the fee line it depends on has been in structural decline: non-interest income has fallen 22% over ten years from a $56 million quarterly peak. Growing wealth fees 11.5% is a genuine achievement; doing it fast enough to offset a maturing margin is the open question.

The Deposit Franchise Under Test

Average deposits fell 0.8% and end-of-period balances dropped $411 million. Part of that was deliberate: management ran off roughly $200 million of expensive public money, echoing July's guidance that 10%–15% of the $2 billion book at 3.5%–4% could go. This quarter it is public funds — the single highest-ranked keyword in the company's own Q3 vocabulary — and roughly “10% of that portfolio still has an opportunity to either reprice or to be run off” — Bradley Satenberg, Chief Financial Officer · 2026-10-26, though not this quarter. The rest of the outflow was ordinary: construction draws, owner distributions, condo project completions.

The tell is what filled the hole. Deposits peaked near $21.2 billion at the end of 2025 and have slipped since, and the bank leaned on short-term borrowings — about $345 million on a spot basis, averaging $120 million of overnight money — a funding line the company had not needed to emphasize in years. Management expects deposit growth to resume modestly in Q4 and again in Q1, which would let borrowings bleed off. Cost of deposits held at 1.27%, non-interest-bearing balances were 27% of the total, and the guided upward beta is 34% over several quarters. If that beta arrives while the deposit base is still flat, the margin arithmetic gets tight.

A Credit Blip, and a Buyback Bid

Asset quality was strong by every measure except one. Delinquencies of 46 basis points, net charge-offs of 9, and criticized assets down 44 basis points to 2.37%. But nonperforming assets jumped to 30 basis points from 8 — a mirror of the NPA ratio theme surfacing across the financial sector this quarter — driven entirely by a single matured commercial real estate loan. Risk chief Brad Shearson was unambiguous that it is idiosyncratic and that “any loss associated with the sale is expected to be immaterial and had already been adequately reserved for” — Bradley Shairson, Chief Risk Officer · 2026-10-26. Add it up: a 49% weighted-average LTV on the residential book, FICO scores near 798, no CRE property type above 10% of loans.

The more interesting signal is capital. Management bought back $20 million of stock in Q3 at roughly $78 per share, plans another $20 million in Q4, and the board lifted the authorization by $200 million with only $69 million left on the old plan — an unusual move for a bank whose shares have since fallen below what it was paying. The CFO's rationale was blunt: “We think it's undervalued.” — Bradley Satenberg, Chief Financial Officer · 2026-10-26 At 1.7x book and 13.6x earnings, and with net interest income having compounded 42% over a decade, that is a defensible claim — provided the organic growth the new CEO promised actually shows up before the repricing tailwind fully fades.