Ask a trader what the transport cost to market in Kenya actually is for one bag, and you get a shrug, or a number that only holds for that trip, that road, that week's diesel price. The honest answer is that the cost is never one figure. It is a chain of separate charges, some fixed per vehicle and some fixed per bag, and the chain only makes sense once you take it apart step by step and see which links move with distance, which move with the number of bags on the truck, and which do not move at all.
Measured by NuaSense weather stations and soil probes on Kenyan farms, over the period stated with each figure. Past readings, not a forecast.
Loading the bag onto the vehicle
Labour comes first. It's fixed per bag, no matter how far that bag goes afterward. A gang loads a bag whether the truck is headed to the nearest collection point or much farther to a regional market. Distance plays no role here at all. That makes it the easiest cost to compare from trip to trip, and the easiest for traders to underquote. It feels too small to matter beside fuel. But with a single bag, it isn't small: the loading labour spread over one bag is the entire cost. Put fifty bags on the same truck and it becomes a fiftieth of that. This is where the per-bag number first starts dropping as load size rises, and it drops before the vehicle leaves the yard.
The boda or the lorry: two different cost shapes
A boda with two or three bags and a lorry with two hundred bags aren't the same trip at different scales. They are different cost shapes. The boda's running cost, mostly fuel and rider time, stays close to fixed whether it carries one bag or three. So its per-bag cost falls fast over a small range, then hits a physical ceiling. The lorry's running cost is mostly fuel, driver time and vehicle wear, and those barely change whether it's half full or fully loaded. Its per-bag cost keeps falling over a much wider range before it plateaus. Choosing between the two comes down to where your load sits on that curve. A handful of bags rarely justifies a lorry. A full harvest rarely suits a boda. The middle ground is where the two options compete directly, and where it pays to work out the decision on paper instead of by habit.
Distance, and the fuel bill that scales with it
Distance behaves the way most people think the whole cost behaves: it scales close to linearly with fuel used, but not with everything else in the chain. Kenya's transport policy documentation ties the country's relatively high transport cost to a mix of vehicle operating costs, longer travel times and road crashes, not distance alone. So the fuel bill is only one part of what distance costs a load. Diesel prices move week to week with global markets and local supply. Any per-kilometre figure worked out today is only good until the next fill-up. What stays true no matter that week's price is the shape: the fuel line item splits evenly across however many bags share the trip. On a longer route, an underfilled truck's per-bag cost gets worse in proportion, because the same fuel bill is divided among fewer bags.
The road itself, and why quality is not the bottleneck it looks like
It's easy to blame a bad road for a high transport bill. Sometimes that's fair. But the World Bank's Transport Sector Memorandum for Kenya calls East Africa's road network relatively good overall, with a trucking industry described as mature and competitive, though it notes some stretches of poor-quality or unpaved surface. If the regional road network isn't the main problem, the same memorandum points to border crossings, port delays, slow travel and limited rail capacity as what actually slows surface transport down. A load moving only within Kenya to a domestic market doesn't face those border or port problems directly. Still, the lesson carries over: a rough final stretch into a rural collection point costs a load in travel time and vehicle wear long before it appears as a fuel receipt. That time cost rarely gets itemised on anyone's invoice.
Where border and port delays sit in a domestic load's cost, and where they do not
Most of a bag's trip from a Kenyan farm to a Kenyan market never crosses a border or enters a port. So it's easy to think this whole cost category doesn't matter to a domestic grower. It does, for two reasons. First, any input that came in by import, fertiliser, a spare part for the lorry, a replacement tyre, already carried the same border and port friction the memorandum describes before it reached the farm. That friction is baked into the vehicle's running cost, even on a purely domestic route. Second, a UNU-WIDER policy brief on transport costs in Africa argues that domestic transport costs could be just as significant as the continent's already high international transport costs, and that domestic costs shape where exporters choose to locate. That argument came from a South African case study of firm location decisions, not from Kenyan agricultural loads. Treat it as a reason to take the domestic leg seriously, not as a number to borrow directly.
Market fees, charged whether the bag sells or not
When the truck gets to market, there's a fee for the right to unload and trade there. Fee schedules differ by county and by market. None of the sources behind this piece publishes a national figure or a standard structure. So a farmer should treat any per-bag or per-vehicle fee quoted secondhand as an assumption to check against the market's own schedule, not as a fixed fact. One thing generally holds, whatever the local structure turns out to be: a fee tied to the vehicle or the load, rather than to the sale price, bites hardest in a season when farm gate prices are weak. A trader who sends a half-empty truck pays that portion no matter how much sits in the bed. That is another point where the per-bag cost falls in direct proportion to how many bags share the trip.
Offloading, the mirror of the first step
Offloading uses the same kind of labour as loading, and it deserves its own line rather than being folded into loading. The two don't always happen at the same rate or for the same price. At a market with a slow queue for offloading space, the truck sits idle. On a lorry, that turns into driver time the owner still pays for even though nothing is moving. This is the easiest step to overlook when a farmer compares quotes from two traders, because it rarely shows up on the quote. It appears later, as a driver charging extra for a longer day, or as a truck that leaves later than planned and misses the best selling window at the market. Ask a trader directly how offloading is handled, and whether it is included in the quoted price. That closes a gap a lot of disputes over the final bill hinge on.
The agent, and what the cut actually buys
The market agent takes a slice for finding a buyer, haggling the price and sometimes for guaranteeing payment. For many growers, this is the murkiest part of the chain. The agent's cut is a percentage, not a flat fee, so it rises and falls with the sale price, not with how many bags you send. That means a bigger load doesn't shrink the agent's charge the way it shrinks loading, market fees and fuel. A bigger load doesn't lower the agent's percentage. But it can give you more room to negotiate that percentage down. An agent moving fifty bags in one deal has less reason to hold out for a high cut than one moving five. The choice of channel (farm gate, open market or agent) is its own question. The previous piece on this site, farm gate, market or aggregator pricing, worked through where the money actually lands once the sale is made.
Non-tariff costs: standards and paperwork that never appear as a transport line
Trade rules are no longer just about tariffs. The Transport Sector Memorandum says requirements now cover traceability of the production chain and sanitary and phyto-sanitary standards for farm products, plus persistent non-tariff barriers more broadly. You won't see any of this on a transport invoice, but it changes what a load needs before it can move: paperwork, certification, sometimes specific packaging. Each of those adds a cost before the truck even leaves the farm. If your produce is headed for export, or for a buyer who demands traceability, these costs can outweigh fuel and labour put together. For a bag going to a domestic wet market, it usually doesn't apply at all. That's another reason the real transport cost depends on which market the bag is actually headed to.
Why the per-bag figure falls as the load grows
Line up the fixed costs and the scaling costs, and the pattern is clear once you write it down. Loading labour, market vehicle fees and part of the driver's time act as fixed costs per trip. Call that total C. If the fixed costs for a trip are C and n bags go on that trip, the cost per bag is C divided by n. Send five bags on a trip with fixed costs of C, and each bag carries a fifth of C. Send twenty five bags on the same trip, and each bag carries a twenty-fifth of C, which is a fifth of what it carried before. The road, the fuel price and the market fee haven't changed. Only n changed. That is the whole arithmetic case for joint transport. It isn't a discount you negotiate; it's division. And any group of farmers can use it if they coordinate one trip instead of sending five separate half-loads.
Organising joint transport without losing track of whose bag is whose
The arithmetic favouring joint transport is simple. The practical objection growers raise is that pooling a load with neighbours makes it hard to track ownership, quality and who pays what share of the fixed cost. You can solve that with paperwork most transport groups already improvise: a shared manifest that records whose bags are whose by weight or count, a pre-agreed split of fixed costs based on how many bags each person contributed, and one contact person who deals with the loading crew, the market fee desk and the agent. That way the negotiation happens once, not five times. A group that already tracks its own harvest weights, the kind of record a farmer keeps from weighing produce before sale, has an easier time splitting the bill fairly than one negotiating from memory. The organising cost is real, and it lands on someone, usually the person who has done this before and is willing to chase the paperwork. That person's time is a cost worth naming, not assuming it comes free.
Licensing friction that used to sit on top of every trip
Before 2014, licensing a Kenyan truck meant dealing with three separate bodies: the police for inspection and driving tests, the Kenya Revenue Authority's Road Transport department for licences, and the Transport Licensing Board for annual cargo and bus licensing. That split created loopholes for cartels to process fake licences and logbooks. Getting a logbook or number plate could take as long as six months. The National Transport and Safety Authority's TIMS digital platform, launched in 2016 with part funding from a $500 million World Bank credit, now handles the same paperwork in three days. The Matatu Transport Vehicles Association says an operator can apply for all necessary licences within five minutes without an intermediary. This isn't a claim that freight rates fell. What it removes is a specific kind of delay and rent-seeking that used to stand between a truck owner and being legally on the road. That was a real cost even though it never showed up as a line on a transport invoice.
Where the cost chain breaks down
The arithmetic above assumes fixed costs stay fixed. They don't always. A Strathmore study on the economic cost of congestion puts a direct cost on Nairobi traffic. A route through or near the capital can turn a predictable trip into one where driver time balloons without warning, and that driver time is the fixed cost the whole per-bag calculation rests on. A breakdown mid-route does the same thing from another angle. The fixed cost is no longer C; it is C plus a tow, a delay and possibly a spoiled load if the cargo is perishable. Kenya's transport policy names vehicle operating costs, travel time and road crashes as drivers of the country's high transport cost. A crash anywhere on the route is the failure that turns a cost calculation into a loss, because a damaged or delayed load can miss the market day it was aimed at completely. To build an honest per-bag sum, you need a margin for exactly these failures. Don't assume the fixed cost from a good day will hold on a bad one.
Working out your own number
Not one of the sources behind this piece publishes a shilling figure for moving a single bag from a Kenyan farm to a Kenyan market. That figure doesn't exist as one national number. It depends on the vehicle, the route, the season's diesel price and the market's own fee schedule, and all of those move independently. What you can do as a grower is build the sum described above using real quotes from a real route. Ask the loading gang, the driver, the market fee desk and the agent what each charges. Add them up as C, divide by the number of bags you expect to send, then repeat the division at a few different load sizes to see where the curve flattens out. That flattening point, the load size where adding more bags barely moves the per-bag cost, is the number worth knowing before the next harvest, not the single average figure a trader quotes from memory. Coordinating harvest timing with neighbours, the kind of decision covered in NuaSense's overview of IoT applications in Kenyan agriculture, is what makes a full truck possible. And a full truck is the only thing in this whole chain that reliably brings the per-bag cost down.
Also drawn on for this piece: Kenya : Transport Sector Memorandum, Volume 1. Strategy; Updated Integrated National Transport Policy.