Eli Lilly Q3 2026: The GLP-1 Price Cut Its Margin Absorbed

Lilly collected 15% less per dose in Q3 2026 than a year earlier, its steepest price drop in five quarters. The worry I keep seeing is that this is the GLP-1 price war arriving, with government deals, Novo Nordisk's price resets, pills and compounded copies pushing the drugs toward commodity pricing. I think the margin says something else. If Lilly's costs per dose had stayed where they were, a 15% price cut would have taken its gross margin to about 81%. It reported 84.4%.

That gap is the whole argument. If Lilly's edge were pricing power, every point of price it gave up would come straight out of margin. If its edge is that it makes these drugs at lower cost than anyone else, and keeps getting cheaper as volume grows, the margin holds while price falls. The second is what the last five quarters look like to me.

The bears are right that the price line is getting worse. Volume growth has still beaten the price decline by roughly 40 to 50 points every quarter, so revenue kept compounding, and the split inside Q3 shows where the cut went.

Most of the worldwide decline is outside the US, where Mounjaro grew 93% to $5.7 billion. The release doesn't say which markets or price tiers drove it, so I won't guess. What matters for the moat is what the cut did to the margin.

Here's the check I ran. Take the gross margin from the same quarter a year earlier, cut the price by that quarter's reported decline, and keep the cost per dose the same. That's the margin Lilly would have printed with no cost improvement at all. The gap between that and the margin it actually reported is what lower unit costs (and anything else, such as product mix) paid for. The margin here is non-GAAP, which for Lilly's gross margin only strips out amortization of acquired intangibles, so one-offs like milestone payments are still in it.

When the cut was small, as in Q4 2025 at 5%, the gap was under a point. When it was 10% or more, the gap ran from 1.6 points in Q1 2026 to more than three points in the other three quarters, and it was widest in Q3 2026, the biggest cut. It doesn't rise neatly with the size of the cut, but a margin that beats the flat-cost line in every quarter looks to me like a cost curve being passed on, which is what a scale advantage should look like.

GLP-1 drugs are peptides, which are hard to make in bulk, and the lines that fill injectable pens take years to build. Lilly says it has announced 10 new US sites since 2020, with new expansions in Texas and Virginia in the Q3 release. Plants like these have high fixed costs and low costs for each extra dose, so the more doses run through them, the cheaper each one gets. Lilly doesn't report cost per dose, so this is my reading of the margin, but it fits: cut price, sell more, spread the fixed cost wider.

Lilly also raised its 2026 revenue guide to $91 to $92 billion from $85 to $87 billion in the same quarter its price fell fastest. A company whose moat was eroding under price pressure wouldn't usually be raising its sales outlook this hard.

The strongest case against me starts with the one-offs. Lilly says Q3 revenue included about $900 million of benefits, including $410 million of milestone payments on Jardiance, a diabetes drug it sells with Boehringer Ingelheim, plus favorable adjustments to rebates and discounts. The release says all three quarters of 2026 had milestones of this kind, and Q2 had rebate adjustments big enough that US price would have fallen about 9% rather than 3% without them. Revenue like that has almost no cost, so it lifts the gross margin.

I tried to size it. Take the $900 million out of Q3 revenue and assume none of it had any cost, and the margin falls from 84.4% to about 83.8%. That's still about three points above the flat-cost line. Lilly says no further Jardiance milestones remain, so Q4 2026 will be the first clean read.

The second objection is that Lilly isn't choosing these cuts. Lilly says the US decline came mainly from Zepbound, and the release's own risk language points to its voluntary pricing agreement with the US government. On this view the cuts are imposed and permanent, and mix and milestones, not scale, have held the margin so far. I think that's partly right about the cause, and it doesn't change the test. Whether a price cut is chosen or forced, the question is who can afford it, and so far the margin says Lilly can.

The third is direction. The price decline tripled in three quarters, and volume growth slowed from 65% in Q1 to 53% in Q3, with Mounjaro's growth falling from 125% to 65%. If the two lines keep converging, there's a point where volume no longer outruns price and revenue growth stalls however cheaply Lilly makes the drug. And if the category moves to pills, which need far less pen-filling capacity, the plants Lilly built are worth less. Novo already sells a Wegovy pill, and Lilly's own, Foundayo, sold $184 million in its first full quarter. I'd count a quarter where price falls faster than volume grows as a real strike against me.

Price pressure doesn't land the same way across drug names, and the framework's moat ratings show why. Regeneron and Vertex are in the table as the contrast: their moats rest on patents and drugs nobody else sells, so they have no volume loop to fall back on if price drops. Hims & Hers sells Novo's branded GLP-1s on a subscription platform, so a cheaper Lilly drug is a rival offer it has to compete with.

Lilly's scale economics are rated strong and Novo Nordisk's intact; the pillar doesn't apply at the other three. Novo is the only other company making these drugs at this volume, so it's the real test of whether scale decides who wins the price war. If its Q3 shows price falling without the margin holding, that's the price war the bears describe, landing on the company with less scale.