Gentex Corporation GNTX

Key data

The business

Gentex makes the small piece of glass you glance into a dozen times a drive: the interior rearview mirror that dims itself when the car behind you turns on its brights. It sells that mirror, and its exterior versions, to nearly every automaker, and owns a dominant share of the auto-dimming mirror market it more or less invented in Zeeland, Michigan. The moat is patents plus scale. It has made electrochromic glass longer and cheaper than anyone, and swapping a mirror supplier mid-platform is not worth the trouble to a carmaker.

The growth story is not more cars, it is more Gentex per car. The Full Display Mirror streams a rear camera feed onto the mirror glass, HomeLink opens your garage door, and newer bets, dimmable windows on the Boeing 787, dimmable sunroofs, medical-imaging and iris-scanning ventures, each add dollars of content to a vehicle. Last year it bought the Premium Audio business, the old VOXX, lower-margin and outside the core, and spent 2026 wringing costs out of it.

The numbers

The quarter that matters is the last one.

Revenue slipped 1% against the same quarter a year ago, yet net income rose 19% and earnings per share jumped from $0.43 to $0.54, beating the Street's $0.50. The reason is margin, and it is the whole 2026 story in one line: flat sales, sharply higher profit, because the cost side finally turned.

Step back and the compounding has been poor. Revenue grew 10% a year across the span, but per-share earnings peaked at $1.84 in 2023 and then fell two years running to $1.74 in 2025, even as revenue set records, as chip-era costs, inflation and VOXX dilution ate the margin. The recovery now under way is the reversal: first-half 2026 earnings of $1.00 a share are up 18% on the same period last year. The margin trend is the tell.

Q2's 37.0% gross margin is the highest in the visible run, a clean step up from a line that had been stuck near 34%. That is also the exact test the margin-versus-units watch was built around, and only half of it cleared: margin held, at its best level in years, while units did not, with volume still leaving revenue down 1%. The market prices this as an ex-growth supplier at the very bottom of its twelve-year range, and the Street models only $2.9B of revenue by 2028, low-single-digit growth. The view here is that the margin turn is structural, VOXX turning profitable, cost programs landing, richer mix, not a one-quarter gift. The print that settles it is gross margin holding 36% or better into year-end while revenue stops shrinking.

Management

Insiders sold and did not buy: thirteen sales worth $2.8M over the past year, none the other way, led by an officer listed as Downing Steven R selling $866K in February with two other officers selling alongside him the same day. Plan status is not disclosed, so read it as ordinary compensation-driven selling rather than a signal, though a cluster of officers exiting on one February date is worth a mental note. Where management has spoken loudly is buybacks: $315M repurchased in 2025 and $206M in 2024, funded entirely from cash with no meaningful debt, at a price near the low of the decade. That is the right time to shrink a share count, and it puts a floor under the stock. Free cash flow runs near $500M against a $4.9B market value, a 10% yield, and earnings have beaten or met the Street in three of the last four quarters.

How it fails or surprises you

Volume outruns content (downside). Revenue fell 1% year over year despite content gains and a full quarter of Premium Audio. Global car production is flat to down, and if Chinese and European automakers design out or in-source mirrors, more dollars per car cannot offset fewer cars. The number that tells you first: quarterly revenue turning positive again against unit shipments.

The margin turn is real (right tail). Q2 gross margin of 37% is the best in the visible series. If it holds in the high thirties while Full Display Mirror and dimmable glass scale, earnings re-rate from 12x back toward the 15x to 19x this stock normally fetches, and the market pays nothing for that today. The print: gross margin at or above 36% for two more quarters.

The bounce is borrowed (downside). The thesis is content growth, yet revenue is flat and 2025 earnings fell outright. If Q2's margin leaned on VOXX integration one-timers or a non-recurring tariff refund rather than a durable cost structure, you own a no-growth supplier correctly priced at 12x and the re-rating never comes. Watch gross margin, stripped of any one-time items, over the next two prints.

Closing thoughts

Two or three more quarters tell you whether the margin recovery is structural. Gross margin holding at 36% or above while revenue stabilizes means the cost structure genuinely changed and the compounder is back. Margin at 34-35% keeps you guessing. Margin below 34% while revenue keeps falling means you own a cyclical at a cyclical price, 12 times earnings. The downside is bounded at that multiple: net cash, a 10% cash yield, a heavy buyback running at the bottom of a twelve-year valuation range. The fatter tail is up because the market pays nothing for the margin turn if it is real, and you are paying so little to find out.

The bet is still that Gentex sells more of its self-dimming mirrors and screens per car while the world builds fewer cars, and that the cost side has genuinely turned. What breaks it is margin fading back toward the low thirties while unit volumes keep falling. The one pair of numbers that tells you first is gross margin against revenue growth. Hold the high thirties while sales stop shrinking, and the compounder is real. Give the margin back while units keep sliding, and you own a cyclical at a cyclical's price.