IBKR's Q3: Customer Margin Loans Surged 36% Right Before the AI Rout Tested Them

Interactive Brokers closed Q3 2026 with $1.91 billion in reported net revenue and a $0.69 GAAP diluted EPS, up 17% from $0.59 a year earlier. Adjusted EPS of $0.67 matched consensus; reported and adjusted pre-tax margins both held at 78%. Net interest income of $1.15 billion (+19% YoY) accounted for the bulk of the beat — roughly60% of total revenue. The board raised the quarterly dividend from $0.0875 to $0.15, lifting the annual run rate from $0.35 to $0.60, a 71% bump that signals management's confidence in the rate-driven NII tailwind.

Strip out NII and the picture is quieter. Customer stock share volume fell 24% year-over-year to 90.2 billion shares, options contract volume slipped 1% to 415,000 contracts, and annualized DARTs per account dropped from 195 to 168 — a 14% decline. Management acknowledged the slowdown on the call, describing \"moderation in client activity.\" The Q3 beat came from heavier bets on bigger days, not from more trading.

The single most important number in the Interactive Brokers release isn't the EPS beat. It's the $105.2 billion in customer margin loans — up 36% from a year earlier and up 4% sequentially from August's $100.7 billion, per the September monthly metrics. The split-month data matters: August margin loans actually fell 7% month-over-month, so the September reacceleration (+4% MoM) shows customers re-leveraging into the final days of Q3, exactly the period the AI trade was peaking. Management called it clients moving to a \"more risk-on\" posture on the earnings call. The bad-debt line reinforces the read: Q3 2026 booked a $1 million recovery, against a charge in the year-ago quarter, meaning the leverage ramp has not yet produced credit losses at scale.

Two structural shifts sit underneath those numbers. First, account growth at 35% YoY is outpacing DART growth at 18% YoY — meaning new accounts are smaller and trade less frequently, dragging down DARTs-per-account. Second, customer credit balances of $186.2 billion grew 20% YoY, signaling that even as customers levered up on the margin side, they kept a parallel cash hoard at the broker. IBKR pays interest on those credit balances — $1.028 billion in Q3 — and the spread between what it earns on the $105.2 billion margin book and what it pays on the $186.2 billion credit pool is the core engine of the 19% NII growth.

Interactive Brokers and Charles Schwab both reported on Oct. 15, and the contrast in how they monetize higher-for-longer rates is the clearest read on each business model. IBKR runs a customer-funded book. Q3 net interest income of $1.15 billion breaks down as $1.114 billion earned on customer margin loans, $815 million on segregated cash and securities, $65 million net from securities borrowing and lending, $225 million in other NII, minus $1.028 billion paid out on customer credit balances. Charles Schwab, by contrast, reported Q3 net interest revenue of $3.476 billion (+14% YoY) and a NIM of 3.04%, up 12 basis points from Q2's 3.00% and up from 2.88% a year earlier. SCHW runs $451.5 billion in average interest-earning assets funded by sweep balances and bank deposits — a fundamentally different plumbing that converts rate moves into margin expansion, not balance-sheet growth.

The implication for an AI-led drawdown: SCHW's NIM expansion is rate-driven and largely insulated from equity-market volatility, though a sharp selloff can pressure sweep-deposit balances. IBKR's NII is volume-driven — it depends on customers continuing to lever $105 billion-plus against the broker. The CME FedWatch tool now prices an 89% probability of a December Fed rate hike, a structural tailwind for both names. The downside scenario is asymmetric: if the AI rout persists, retail shrinks margin balances and IBKR's NII growth decelerates; if customers reload on dips, the +36% YoY margin book compounds further.

The Oct. 15 print cleared the bar. The Oct. 16 tape cleared the room. Interactive Brokers closed Thursday at $86.03 (+2.25% on the day), slipped to $85.52 after-hours, then fell 3.88% Friday to $82.69 — a three-day net of roughly -2.2% from the pre-print close. Charles Schwab dropped 2.34% on the same session. The Friday tape was a clean AI-leverage unwind: the Nasdaq Composite fell 2.14% to 26,140, the S&P 500 fell 1.09% to 7,659, and the Dow slipped 0.34%. For the week, the Nasdaq lost 4.5% — its worst week since a four-week winning streak snapped.

Here is where the brief's framing of \"first pure read on whether retail bought the AI dip\" needs an honest correction. Q3 2026 ended Sept. 30, and the Interactive Brokers press release covers July-September trading only. The Oct. 16 selloff came after the print, so Q3 is not the post-rout read — it's the last clean pre-rout read. The relevant question is what IBKR's September monthly metrics showed about positioning heading into October: margin loans +4% MoM, accounts still growing at 35% YoY, credit balances steady. Customers went into the AI trade's rollover levered longer than the prior year and not yet showing bad-debt losses. Whether they bought, sold, or held through Friday's 2.14% slide is the next data point — and it shows up in the October monthly metrics, due in early November.

The next retail-only window is Robinhood, reporting Q3 on Oct. 27 — ten days after the AI rout. Robinhood's funded-customer count, net deposits, and options-trading metrics will be the cleanest read on whether retail treated the Oct. 16 slide as a buying opportunity or a margin call. The transmission is direct: if HOOD shows rising fund balances and active options, the Q3 Interactive Brokers leverage ramp reads as a dip-buying setup; if it shows shrinking balances, the +36% YoY margin loan growth was a pre-rout mistake.

The honest synthesis: Interactive Brokers's Q3 was a structurally strong print — net interest income up 19%, customer accounts up 35%, dividend up 71% — built on a customer base that was demonstrably levered into the AI trade's final weeks. The Oct. 16 rout hasn't yet been fully priced into IBKR's fundamentals because it happened after quarter-end. The risk for the rest of Q4 is that the +36% margin loan growth unwinds as forced deleveraging; the upside is that midterm-election volatility reactivates trading and the leverage book compounds. Either way, the 78% pre-tax margin has very little room before the model breaks — and management's 71% dividend raise signals that they believe the rate tailwind will outlast any AI-trade correction.