C.H. Robinson Worldwide, Inc. CHRW

Key data

The business

C.H. Robinson is a freight broker: it owns no trucks, it matches shippers who need something hauled with carriers who have space, and keeps the spread. Two segments. Transportation was $4.52B of the $4.93B booked in Q2, spanning North American truckload and less-than-truckload plus Global Forwarding's ocean and air lanes; Sourcing, mostly produce, was the other $409M. The model is nearly asset-light: capex ran $20M on $16.2B of FY2025 revenue. The moat is the largest carrier network in North American brokerage plus decades of pricing data, which lets Robinson quote and cover a load in minutes when a shipper's own truck falls through. Headline revenue swings violently with truck rates, from a $24.7B peak in 2022 down to $16.2B in 2025. What matters is the spread it keeps, not the gross freight dollar, so the cycle bruises the top line far more than the profit.

The numbers

The story is a profit recovery running well ahead of the revenue recovery.

Q2 2026 was the inflection: revenue up 19% year over year as truckload volumes and rates firmed, net income up 22%, the fourth straight quarter beating estimates on adjusted earnings. Personnel expense barely moved, $335M to $338M, while revenue rose a fifth. That flat cost base against a rising top line is the entire case.

Earnings troughed at $2.72 in the 2023 freight recession and have compounded to $4.83, with 1H 2026 at $2.78 already annualizing above $5.50, all while revenue sits a third below the 2022 peak. FCF converted at roughly 1.5x net income last year (about $895M on $587M), because capex is a rounding error. That cash bought back stock and covers a $0.63 quarterly dividend.

Here is the tension. Net revenue margin, the cut kept on each freight dollar, compressed from 16.8% to 15.0% as purchased transportation rose faster than what Robinson could charge. Yet operating margin on that net revenue expanded from 31% to 35%, because the cost of arranging it barely grew. Q3 will test whether this is plateau or peak: a thinner cut, converted harder. The variant the market underweights is that this conversion is partly structural, not just cyclical leverage, which lifts through-cycle earnings power above what the 2023 trough implied. The print that settles it is Q3 net-revenue margin against flat personnel.

Management

Insiders are net sellers, and heavily: $18.9M sold against $502K bought over twelve months, led by Freeman ($5.8M, Nov 2025) and Castagnetto ($6.9M across two Nov 2025 sales); plan status is not disclosed on these Form 4s, so read it as officers trimming after the stock touched $210 and now sits at $147. Capital allocation is the more honest tell. The company repurchased $354.7M in FY2025 and another $212.7M in Q1 2026, cutting the share count to 116.9M as of July 29, 2026, funded entirely from free cash flow rather than debt drawn for the purpose. Pay is built around the LEAN productivity program that is producing the flat-headcount profit growth. The record reads consistently: buy the stock, including near today's price, while the people running it sell their own.

How it fails or surprises you

Net-revenue margin keeps bleeding. The cut per freight dollar already fell 180 basis points year over year to 15.0%. If a sharp rate spike squeezes it below 14% in Q3 2026 while cost stays flat, the operating leverage reverses fast; this is the fact the recovery story explains least, and the one that would prove it wrong.

The freight cycle rolls over again. Revenue is still a third below the 2022 peak, and Global Forwarding rides tariff-exposed ocean and air volumes. A renewed volume contraction, especially in forwarding, drops net revenue faster than headcount can follow, re-running 2023.

A real up-cycle lands on a flat cost base (right tail). If truckload rates and volumes both recover through 2027 with the LEAN cost structure intact, net revenue drops to the bottom line at 30%-plus incremental margins. The market pays 19x EV/EBITDA for a steady compounder, not for a cyclical earnings surge, so a run toward $7 of EPS is not in today's price.

Closing thoughts

This is a name where a specific print resolves the debate rather than one already settled by the tape. On the other side sits a market paying near the top of a twelve-year range, 19.3x EV/EBITDA against a 13.9x peer set, for what it treats as a quality asset-light compounder. My read is that Q2 showed something the multiple does not fully credit: cost per shipment that no longer rises with volume, which raises the earnings the business throws off across the whole cycle. Q3 net-revenue margin and personnel expense convert that from claim to fact. The left tail, a margin squeeze into a cyclical downturn, is the fatter near-term risk and the one to size against; the right tail, an up-cycle on a fixed cost base, is worth more but needs the freight market to cooperate.

The bet is still that Robinson keeps standing between the companies that need freight moved and the truckers who haul it, keeping a cut of each load while paying fewer people to arrange it. What breaks it is the cut shrinking faster than the cost base does. The pair that tells you first, every quarter: net-revenue margin against personnel expense. If margin drops below 14% while headcount cost climbs, the thesis is wrong.