Freight Derivatives and FFAs

Freight is the price of moving cargo, not the price of the cargo itself. That distinction matters: a dry-bulk vessel is a mobile, capacity-constrained production asset whose supply responds slowly, while demand depends on trade routes, port congestion, commodity flows, and voyage duration. Forward Freight Agreements (FFAs) let market participants trade published route indexes without owning ships. For quants, they are a compact case study in a non-storable service with extreme cyclicality.

What an FFA settles on

An FFA is a cash-settled agreement on an average freight index for a defined route, vessel class, and delivery period. A long benefits when the realized index exceeds the contract rate. Route details are the contract: Capesize and Panamax exposures differ, and a route may quote dollars per day or dollars per tonne.

Contract dimension	Economic exposure
Vessel class	Cargo capacity, fuel use, port compatibility
Route	Origin, destination, canal and ballast effects
Quote unit	Time-charter equivalent or voyage rate
Settlement month	Average published index over that month
Clearing venue	Counterparty, margin, and liquidity profile

Do not treat a route index as a universally hedgeable shipping beta. A charterer may face a different vessel, port delay profile, or bunker cost. The hedge is valuable only to the extent that its settlement index covaries with the actual future freight bill.

Supply-demand mechanics

Fleet supply is largely predetermined over the next several quarters: vessels are ordered years ahead, scrapping is slow, and speed changes add only a limited short-run margin. Demand is vessel-days, not just cargo tonnes. Longer routes, port waiting, weather, and ballast repositioning can consume capacity even when cargo volumes are unchanged.

effective_supply ≈ fleet_dwt × utilization / voyage_duration
freight_rate = g(cargo_tonnes, effective_supply, congestion, fuel_cost)

This produces a convex rate response near full utilization. When spare ships are scarce, a small demand increase can create a large spot-rate jump. Conversely, newbuild deliveries can depress rates for years. A linear model of monthly rates will tend to understate both tail risk and mean reversion after capacity additions.

Curves and seasonal signals

The FFA curve embeds expected seasonality and the market’s view of the shipping cycle. Compare a contract with a conditional fair value, not with a trailing spot average. For dry bulk, grain exports, iron ore flows, Chinese holidays, and weather can all shape seasonal route demand. Tanker markets add refinery runs, floating storage, and geopolitical routing effects.

import numpy as np

def route_fair_value(features, coefficients):
    x = np.array([1.0, features['iron_ore'], features['congestion'],
                  features['fleet_growth'], features['bunker_cost']])
    return x @ np.asarray(coefficients)

def zscore(value, mean, std):
    return (value - mean) / max(std, 1e-6)

A more robust model uses separate state variables for demand, fleet availability, and port friction, with a nonlinear mapping to rates. Evaluate forecast calibration by delivery month and regime. A model that forecasts quiet markets well but misses scarce-capacity states may have unacceptable short exposure even if its RMSE looks good.

Relationship to commodities

Freight is intertwined with commodity arbitrage. A high ocean rate can close the price gap between two regional iron-ore or grain markets, while a low rate opens it. But shipping derivatives and commodity futures hedge different legs. Combine them only after matching route, cargo timing, and currency. The curve methods in commodity spread trading are useful, but a freight spread has an additional operational layer: the physical voyage may not occur on the paper contract’s schedule.

Bunker fuel adds another basis. Depending on charter type, fuel cost may sit with the owner or charterer; route indexes can incorporate an assumed bunker price. A seemingly clean freight trade may therefore be an unrecognized fuel position.

Data and execution issues

Index methodologies, publication times, route revisions, and holiday calendars belong in the dataset. FFA liquidity is concentrated in benchmark routes and near maturities; marks in distant contracts can be dealer indications rather than executable prices. Backtests should apply conservative bid-ask, limited participation, and roll constraints.

Margin is another constraint. Freight can gap during port disruptions, sanctions, or commodity-demand shocks. Since FFA liquidity typically deteriorates in exactly those states, size positions from stressed variation-margin needs. Use a liquidity haircut rather than assuming positions can be closed at the last published index.

Key takeaways

• FFAs cash-settle on a specific shipping route and vessel-class index; contract definition drives basis risk.
• Freight rates are convex near full utilization because vessel supply adjusts slowly.
• Model vessel-days, congestion, fleet deliveries, and voyage distance—not cargo volumes alone.
• Freight and commodity prices interact through arbitrage but require separately matched hedges.
• Contract-level liquidity, marks, and margin are essential parts of an FFA backtest.