The smallholder problem
A 2 ha block, a 5 kg seed bag, and the cost math that makes smallholder cover cropping harder than estate cover cropping.
There is a story the cover crop industry tells about smallholders. It goes: the science is settled, the species are well known, the agronomy is straightforward, so the bottleneck must be awareness or training. Once we educate smallholders about cover crops, they will adopt cover crops, and tropical soils will improve.
I have been to enough smallholder blocks in Pahang, Perak, and Johor to know that this story is partially true and substantially wrong. The bottleneck is not awareness. Most smallholders I have met know what cover crops are. Many have seen a neighbour try them. The bottleneck is economics, and the economics of cover cropping on small acreage are harder than the economics on estate acreage in ways that the agronomy literature rarely engages with.
This is the smallholder problem as I see it.
## What "small" means here
Malaysian smallholder oil palm operations are typically 2 to 25 hectares, owner-operated or family-operated, often without a dedicated agronomy staff. They sit alongside large estates that run thousands of hectares with full agronomy teams, labour rotations, and chemical purchasing departments. The two operate in the same crop but in fundamentally different economic systems.
The cover crop agronomy was developed for estates. It assumes a planting schedule, a maintenance regime, a labour pool, and a multi-year investment horizon. Drop that same agronomy into a 4 ha smallholder block and several things break at once.
## Break one: minimum order quantities
The cover crop seed trade runs at estate scales. A 25 kg bag of MB seed makes sense if you are planting 30 hectares. It does not make sense if you are planting 4 hectares and need 5 kg.
In practice, smallholders end up either splitting bags with neighbours, which works only if neighbours are planting at the same time with the same species, or buying small repacks from local dealers who do not always document the original seed lot. The repack chain is where most quality information gets lost. A 1 kg bag in a smallholder's hand is often two or three hands removed from the certificate, if a certificate ever existed.
We sell smaller packs deliberately. Five hundred grams, one kilo, five kilo. With the original batch germination data attached to each pack. This is not because the unit economics are great at that size, they aren't, but because if a smallholder is going to invest in cover crops, they deserve to know what they bought.
## Break two: labour
The MPOB establishment standard for MB, six-week nursery seedlings, 320 per hectare, two per planting point, works at estate scale where there is a nursery and a transplanting crew. It does not work at smallholder scale. A 4 ha smallholder is not running a six-week nursery for 1,280 MB seedlings.
The options collapse to two. Buy ready-grown seedlings from a third party, which the budget often cannot absorb. Or switch to a broadcast species, CM, PJ, CC, that does not need nursery handling. Most smallholders, sensibly, switch.
Year-one labour for MB-style maintenance, monthly palm-circle walks to keep cover off young palms, is a real cost on an estate and a sunk cost on a smallholding because the smallholder is already doing those walks. The break is in the other direction: if the smallholder does not have the time to walk the rows, MB is the wrong species and the recommendation should be a less aggressive cover. The literature defaults to MB. The right answer for many smallholder blocks is not MB.
## Break three: the validation window
This is the cruel one. The agronomic benefit of a cover crop shows up on a 24 to 36 month horizon. Soil organic matter changes are detectable but slow. Yield effects on the palm itself, if there are any, are even slower. The smallholder is making a year-one cash investment to see a year-three soil benefit, and the year-three benefit is hard to attribute back to the cover crop without a controlled comparison.
Estates can run controlled comparisons. Smallholders cannot. A smallholder who plants cover crops and sees a decent palm year three has no way to know whether the cover crop did anything, or whether the rainfall pattern was just better that year. The feedback loop that would let a smallholder learn from their own cover crop investment is broken by design.
What this means in practice: smallholders who try cover crops once and do not see an obvious effect rarely try again. The species choice and the agronomy may have been right. The reading of the result is what fails.
## Break four: the gap between species choice and species availability
In a perfect world, a smallholder would walk into a local agricultural supplier and find five tropical cover crop species, each with documented quality data, in pack sizes that fit their acreage. That is not what most Malaysian local suppliers stock. The most commonly stocked species at smallholder scale is whichever one moves through the supply chain that month, often CM, sometimes PJ.
If a smallholder asks for Centrosema pubescens and the local supplier does not stock it, the smallholder buys what is available rather than not buying anything. The species choice gets made by the supplier's purchasing decisions, not by what the block actually needs.
We try to close that gap by direct shipping in small packs. It is a partial fix. The real fix would be a denser network of small-pack-friendly agricultural suppliers across the smallholder belt, and that is a structural problem that one seed company cannot solve alone.
## What I tell smallholders when they ask
When a smallholder calls and asks for a cover crop recommendation for their block, I work backward from three questions before I get to species.
How much time do you have to walk the rows in year one. If the answer is "almost none," MB is out. Go with PJ, CM, or a low-supervision mix.
Do you have a way to validate that the cover crop is doing what it is supposed to do. If the answer is "no," set a different expectation. The cover crop is buying you erosion control, weed suppression, and a slow organic matter trend. It is not buying you a visible year-two yield miracle. Plan and budget on the modest expectation.
Can you afford one bag, two bags, or three. Once I know the seed budget, the species choice becomes a constrained optimisation rather than a textbook recommendation. Sometimes the right answer is a half-hectare cover, not a four-hectare cover. A successful half-hectare establishment can teach the smallholder what cover crops actually do on their soil, and then the second year is informed by their own data instead of a textbook.
## What the industry could do better
I am not going to dress this up. The cover crop industry, including me, has been structurally biased toward estate customers because estate customers buy more seed per order. The smallholder segment is harder to serve, lower-margin per transaction, and structurally underserved by everyone in the chain.
The honest version of the cover crop story for smallholders is: the science is correct, the species are real, the agronomy is sound, and almost none of the standard recommendations are built around your constraints. If you make the cover crop decision with the constraints in front of you rather than the textbook in front of you, you will get a better outcome.
If you are a smallholder considering cover crops on your block, the entry point is often a small bag of Calopogonium mucunoides on a half-hectare to see what happens, before committing the full acreage. I will tell you that even if it loses me a bigger sale, because the alternative, buying for the whole block and watching it underperform on the wrong species, is worse for both of us.
You can read more about how we work on the Chemiseed About page at chemiseed.com/pages/about.
Need help selecting cover crop species for your block? Reach out at shaharil@chemiseed.com or visit chemiseed.com/pages/contact.