American Road #4: They Cut Inventory 17%, and It Sold — Checking the Q2 Math

Last time I wrote about why Harley's Deadwood is $2,000 cheaper — a bike built by stopping the habit of adding equipment. The quarterly numbers are out now, so let's check the work. In the second quarter of 2026, Harley-Davidson's North American retail sales rose 3% year over year to 29,751 units. Globally, retail was up 1% to 42,500. On paper that is not spectacular. But for a company whose showrooms had been sliding for years, selling more than the year before is not a small fact. What I want to know is where that increase came from. Was it manufactured with discounts, or did the structure of how they sell actually change? The quarterly materials answer that pretty plainly.

Here is the interesting part. In the same quarter, revenue fell 6% to $1.23 billion and operating income fell 32% to $76 million. Revenue and profit both down, yet more bikes leaving showrooms than a year ago. The answer is inventory. Global dealer inventory was down 17% versus the end of the same quarter last year — down 15% in North America and 24% outside it. More than 85% of quarter-end stock was 2026 models. Throttle back what the factory pushes onto dealers and wholesale revenue obviously drops. In exchange, two-year-old "new" bikes disappear from dealer floors. In Oakland, until recently the same model could differ by thousands of dollars between shops. That looked like an attractive discount, but it was really just proof that there was too much inventory. The company says domestic dealer profitability is expected to double in 2026. The way I see it, this is not a story about changing how they sell. It is a story about changing how they wholesale. Let me add one more figure. HDMC — the motorcycle segment itself — posted revenue of $848 million, up 9% year over year. The 6% decline at company level comes from other pieces, financial services included; the business of selling motorcycles actually grew. Blend those together into "revenue is down, so they're finished" and you have misread it. Cutting inventory 17% also means, in plain terms, slowing the factory. Full-year wholesale guidance now sits at the same 133,500–138,500 units as retail. Build only what you ship, ship only what you build. Sounds obvious, right? Harley has not managed it for a decade.

The other thing you cannot skip is tariffs. Harley booked $45 million in the first quarter and expects $75–90 million for the full year. In the second quarter it took $22 million in tariff costs, largely offset by a $20 million recovery. That offset is partly timing rather than strategy, so there is no guarantee the next quarter looks the same. Even so, the company raised its full-year outlook: global retail units to 133,500–138,500, and HDMC operating income to $10–50 million, up from a prior range of negative $40 million to positive $10 million — from red into black. CEO Artie Starrs says the company made meaningful progress against its Back to the Bricks initiatives. The facts are that inventory and retail both improved. The equally factual counterweight is that operating income is still down 32%. My own read: this is not a comeback. This is the stage where the cleaning of a messy room has finally finished.

So, back to Deadwood. What Back to the Bricks puts forward is the "blank canvas" motorcycle. That bike did not have equipment stripped out of it; it had empty space sold into it. In the scene over here, the bars, seat and exhaust are usually different a week after purchase, and the factory kit goes back in its box the same day. If that is the norm, not fitting it in the first place is cheaper for the maker and the buyer alike. And it is not only Harley. With CFMoto's Ibex 950 landing in North America at 112 hp and $12,499, the persuasive power of the sticker itself carries more weight than it used to. One note for readers in Japan: North American dealer inventory returning to normal means fewer nearly-new and used bikes flowing out of that channel. For two years, that North American overstock quietly pushed Japanese used prices down across the Pacific. Grey-import prices will probably tighten from here, so if you plan to buy, assume the clock is running. What I will watch next quarter is one thing only — whether they can rebuild margin without leaning on price increases. If they hold the black-ink outlook without a tariff recovery propping it up, that is the point where you stop calling it cleaning and start calling it real progress.