Contracted freight rates diverge from the published spot index, and the divergence widens as spot moves. Contract reset timing matters more than the index level: a shipper's effective rate reflects the market of the quarter its contract was signed, not the market this week.
What do freight indices measure?
Spot rates, usually on major lanes, usually for standard equipment. That is a real market and a small share of the volume that actually moves, most of which travels under contract.
An index therefore tells you what the marginal container cost this week. It does not tell you what a shipper is paying across their book, which is the number a diligence needs.
Equipment and lane composition matter as much as the spot-contract split. An index built on standard dry containers on headhaul lanes describes a specific slice of the market, and a shipper whose volume is reefer, out-of-gauge or backhaul is exposed to a different price series entirely. Two businesses can both be correctly described as freight-exposed and move in opposite directions.
How far do contract rates diverge?
Widest immediately after a spot swing and narrowing as contracts reset. In a stable market the two converge; in a volatile one the gap can be the single largest error in a logistics model.
Direction of the divergence flips depending on when contracts were signed relative to the swing, which is why a single adjustment factor does not work.
The asymmetry of the gap is what makes it dangerous in a model. When spot rises, contracted shippers look advantaged and the advantage decays as contracts reset; when spot falls, they look disadvantaged and the disadvantage decays the same way. A model that captures the level but not the decay will be wrong in the same direction for as long as the cycle runs.
When do freight contracts actually reset?
Annually for most large shippers, but staggered rather than synchronised. Two shippers on the same lane can be paying materially different rates in the same month simply because they contracted in different quarters.
That staggering is also what makes a portfolio of contracts less volatile than the spot market, which is a point in favor of contracted books that models rarely capture — until the contracts come up for renewal.
Reset timing is also knowable in advance, which makes it one of the few things in a freight model that can be established rather than assumed. Procurement teams run to a calendar, and the calendar is something they will describe. Knowing when a book reprices converts a rate assumption from a guess into a schedule.
Who can tell you the rate?
Freight procurement staff at shippers, and account managers at carriers and forwarders. Neither quotes numbers freely; both will, on a screened call, describe where a rate sits relative to the market and when it was struck.
Forwarders are frequently the most useful single source because they see many shippers on the same lane and can characterize the spread.
Forwarders come with a caveat worth applying. They see the spread and they also participate in it, so a forwarder describing the market is describing a market they price into. That does not make the account unusable — it makes it worth pairing with a shipper's procurement view, which carries the opposite incentive.
How should a model use spot indices?
As a direction, with contract lag applied. Flowing spot straight into revenue overstates volatility in both directions, because the book being modeled does not reprice that fast — the same error that misreads where margin sits along a food chain.
The lag itself is estimable from the contract calendar, which means this is a correctable error rather than an unavoidable one.
Where the calendar is not obtainable, an approximation beats ignoring the lag entirely. Assuming an evenly staggered book that reprices over twelve months is wrong in detail and far closer than flowing spot through directly, and it has the advantage of being an assumption someone can challenge and improve rather than an error buried in the mechanics.