TL;DR the short version
This week I am in Oslo for the Nordic-African Business Summit, taking part in a discussion about what it will take to build a bankable aquaculture sector in Africa. There is a clear benchmark comparison hanging over that discussion: salmon.
Salmon aquaculture is one of the great industrialisation stories of modern food production. It is therefore tempting to look at tilapia in Africa and ask why we have not managed to do the same thing (yet). The comparison is important, but I think we usually draw the wrong conclusion from it.
Imagine two fishermen. Olaf lives on the Norwegian coast in the late 1960s. Manuel fishes kapenta on Lake Cahora Bassa in Mozambique today. Both are entrepreneurial. Both understand fish. Both work extremely hard. Both can see the attraction of farming fish rather than fishing for them.
But Olaf encounters aquaculture while surrounded by an increasingly sophisticated economic system: organised fisheries markets, roads, ports, public research, social insurance, functioning banks and a government actively trying to preserve economic life in coastal communities.
In contrast, Manuel encounters it while much of that system still has to be built. If he wants to farm tilapia, he needs fingerlings that may not be available locally, feed that may have travelled hundreds or thousands of kilometres, working capital that is difficult to borrow, and a route to market that may require him to build the cold chain himself.
The lesson from salmon is therefore not simply that we need better fish farms. It is that industries scale when the systems around entrepreneurs allow them to specialise. The challenge for African aquaculture is to build that system.
And there is another part of this story that I think we routinely underestimate: Manuel may actually have more entrepreneurial capacity around him than Olaf did. It is simply organised differently. His wife may trade fish. His children may help with the farm. His relatives may provide labour. His community may create an association.
From what I can see, the unit of African entrepreneurship is not the individual. It may be a network. That should fundamentally change the way we think about building an African aquaculture industry.
Two fishermen
To illustrate the difference between the salmon farming industry in Norway and the tilapia farming industry in Mozambique, I want to introduce you to two hypothetical fishermen.
First, meet Olaf.
Olaf is a composite of the sorts of fishermen living in Norwegian coastal communities in the 1960s, immediately before modern salmon farming began to take off.
Now meet Manuel.
Manuel is a composite of fishermen we meet in the communities around Lake Cahora Bassa in Mozambique, where we have built Chicoa Fish Farm.
The point is not to pretend that every Norwegian fisherman was Olaf or every Mozambican fisherman is Manuel. But the point is to think about the world each man experiences, because incentives matter; and circumstances shape incentives.
Olaf goes fishing
Olaf lives somewhere along the Norwegian coast.
His father fished. His grandfather probably fished as well. His household might combine fishing with some small-scale agriculture or other seasonal work.
Olaf is not rich. Norway in the 1960s is not the extraordinarily wealthy petrostate we think of today. North Sea oil production has not even begun.
But Olaf is living in a country that has spent the two decades since the Second World War deliberately building institutions that show up in very ordinary parts of his life.
When Olaf returns with his catch, he isn’t entirely alone in negotiating with whoever happens to be standing on the beach.
Norway had begun organising fishermen’s first-hand sales decades earlier. By Olaf’s time, the Raw Fish Act gave fishermen-owned sales organisations a legally protected role in first-hand fish sales and the setting of minimum prices. In 1964, negotiations between the state and the Norwegian Fishermen’s Association were formalised through a main agreement.
This is critical. Olaf does not have to spend quite as much time worrying about inventing a fish market. He will worry more about catching enough fish, and getting the right price for it.
Olaf also lives in a welfare state that has been expanding through the post-war period. His children attend school. There is public healthcare. There are systems of social protection. Roads, electricity, ports and communications have been spreading into communities which, only a generation earlier, were far more isolated.
Olaf still has plenty to worry about: the sea is dangerous; a bad season is a bad season; boats need constant repairs; the catches can be erratic; his life is at risk.
But his risks have some boundaries around them.
If his child becomes sick, it is principally a family tragedy and a medical problem. It is much less likely to become a financing event that requires selling his productive assets. If his brother becomes sick, Olaf will care deeply. He might help him. But his brother exists within the same broader institutional system.
There is a state standing somewhere behind both of them.
Olaf can complain about that state, and I am fairly sure that he will. But he can also expect it to function.
Manuel goes fishing
Manuel lives beside Lake Cahora Bassa, a hydroelectric dam on the Zambezi river system in central Mozambique.
His fishing day actually begins at night. Kapenta are attracted using lights, and the fishing rigs operate after sunset. Manuel works comes home in the morning with something extraordinarily perishable, in a place where daytime temperatures can move rapidly above 30 degrees.
Norway’s climate does not remove the need for a cold chain. Fresh fish has always required careful handling. But heat makes life difficult.
Around Cahora Bassa, access to electricity, refrigeration and aggregation remains a real developmental issue – even for Chicoa Fish Farm.
So Manuel lands his catch and immediately sets up his fish to dry in the sun. It loses around 70% of its body weight in the process, which is why he must catch so much of it. That is also why the kapenta industry is in decline in Cahora Bassa.
But if he catches tilapia in the process, he takes it home to eat, or he trades it for fruit and vegetables in the village.
There is no one coming to Cahora Bassa to buy fresh fish. The market for tilapia does not yet exist, and the route to the market doesn’t exist either.
Manuel’s household might look very different to Olaf’s.
Again, I want to avoid caricature. Mozambican families are no more uniform than Norwegian families.
But economic responsibility in Mozambique is frequently organised through much wider kinship networks. When formal insurance mechanisms don’t exist, your family becomes the insurance mechanism.
Manuel may help his parents. He may help a brother. His wife may help her sister. A nephew may need school fees. He will always contribute to funerals within the community and in his extended family.
This gets described occasionally as a burden on African entrepreneurs, but that misses half the story.
The extended family is both an obligation and an institution.
It is a social security system.
Where Olaf can externalise some of his household risks into institutions, Manuel internalises institutional functions into the household.
That distinction is important when someone comes along with a bright idea for an alternative to wild-capture fishing.
“Have you ever thought about farming fish?”
Modern Norwegian salmon farming emerged from remarkably ordinary experimentation.
The famous story is of Ove and Sivert Grøntvedt putting Atlantic salmon into a floating net pen at Hitra in 1970. But they were part of a wider period of experimentation by small coastal operators. The Norwegian Fish Farmers’ Association was founded that same year. Salmon farming spread quickly enough that, by 1973, Norway had introduced legislation specifically designed, among other things, to keep ownership local and ensure that the new industry benefited coastal communities rather than simply being captured by large industrial groups.
So imagine Olaf hearing about it. Perhaps somebody at the quay is talking about a farmer further down the coast, there is something in the newspaper, or his cousin has seen a cage.
Fish being raised as livestock in the open sea.
It may have sounded strange, but also strangely logical.
Olaf may not immediately sell his boat, mortgage his house and become a salmon farmer.
But he is interested. Then maybe he helps someone, or perhaps he invests in a cage with a relative.
Maybe he eventually applies for a licence.
And critically, as the idea gains momentum, institutions begin accumulating around him.
Publicly backed research starts solving problems which would be absurd for an individual farmer to solve himself. In 1971, the publicly funded AKVAFORSK began the systematic salmon breeding programme that eventually became foundational to modern farmed salmon genetics.
Government policy increasingly treated aquaculture as regional economic development. Working-capital finance in the developing industry was supported through government guarantees. Later assessments of the Norwegian experience specifically identify guarantees enabling farmers to borrow operating capital from local commercial banks as an important part of the industry’s growth.
Private entrepreneurs still took risks, of course. They still lost money. They still failed.
But the important point is that Olaf did not have to privately invent every institution required to make salmon farming possible.
Somebody else could work on genetics. Somebody else could manufacture feed.
This is what industrialisation delivers: the ability to specialise. The more synergistic industrial functions that exist, the easier it is to focus on what you’re good at.
Now offer Manuel the same opportunity
Tell Manuel that he should start farming tilapia. He likes the idea. Of course he does: kapenta fishing is physically demanding and uncertain, and catches are declining. A cage full of fish that stays in roughly the same place has an intuitive appeal.
Then he starts asking questions.
Where do I get the fingerlings?
This is not a minor detail.
The availability and quality of fingerlings remains an identified constraint to small-scale aquaculture in Mozambique and across Southern Africa.
Fortunately, Manuel can buy them from Chicoa.
But how does he transport them alive? Who pays for that? How many does he need?
What happens if there is a shortage precisely when his cage is ready? Then:
What do I feed them?
Feed is the largest input cost in tilapia farming. Mozambique does not have the mature aquafeed manufacturing infrastructure of established aquaculture economies, and access to affordable, quality formulated feed remains one of the sector’s principal constraints.
Can he buy one bag?
Or does somebody need to import a truckload?
Does he pay cash?
Can he get credit?
If feed prices rise halfway through the cycle, what happens then?
And then comes the question that is routinely left until far too late in agricultural development projects:
Who buys the fish?
Where? At what size? At what price? Who harvests them? Do I need ice for that, and if so, how can I get some here before it melts? Who owns the truck? Will the buyer pay immediately?
And while Manuel is working through all of this, there is a bigger problem.
His fish need to eat every day. His children do too.
Fishing produces irregular but frequent cash, while aquaculture consumes cash for months before producing revenue.
So he decides to keep a fish pond with a few fish in it, that he can consume himself. Especially if the fingerlings are cheap.
Olaf was experimenting with a new method of producing fish.
Manuel may be experimenting simultaneously with hatchery access, feed distribution, veterinary support, working-capital finance, logistics, cold storage, formal markets and household liquidity.
Then we look at the failure rate and conclude that small-scale African aquaculture is difficult.
Well, yes it is.
The anchor farm changes the equation
This is why I keep coming back to the importance of anchor commercial farms. A large farm is sometimes presented as the alternative to small-scale agriculture – but I think that this is exactly backwards in an early-stage industry.
A properly designed anchor farm creates the infrastructure that makes smaller producers possible.
This was our original thesis when we built Chicoa.
A commercial farm needs fingerlings so it builds a hatchery. It needs feed anyway, so it creates enough demand to justify regular truckloads, warehousing and eventually local manufacturing.
It needs to sell fish, so it builds relationships with traders, retailers and distributors. It needs cold storage, veterinarians, cages and nets.
It needs people who understand nets and boats and fish health.
And because a commercial farm cannot afford to run every input down to zero, there is naturally spare capacity around the edges.
That spare capacity is where an industry begins.
Suddenly Manuel does not have to build a hatchery: he just needs to buy fingerlings.
That is a completely different economic problem.
He does not have to import a truck of feed: he needs ten bags.
He does not have to create the market for tilapia: he needs access to an offtaker.
The anchor farm turns industries into live transactions.
And transactions are things entrepreneurs can work with.
This is exactly why, in our experience, you find small fish farmers emerging around established commercial farms. They could not economically justify creating the infrastructure. Once somebody else has created it for commercial reasons, they can access that infrastructure at marginal cost. It is something I wrote about a few years ago when trying to figure out what an actual theory of industry development might look like.
Seed. Feed. Training.
But what we really mean is:
Seed, feed and training that are affordable, close by, and available at the moment they are needed.
That is a critical qualifier..
One more thing: Manuel isn’t Olaf
There is another consideration in this story.
When we list all Manuel’s constraints, we make him sound like a helpless beneficiary.
He isn’t – quite the opposite. He is always looking for new ways to make money. He will try most things. But success does not mean the same thing for Manuel as it does for Olaf.
Imagine that Manuel’s first cage works. Then the second one.
His wife notices that selling the fish to the first buyer is leaving money on the table, so she starts taking some into the local market. His eldest son helps with feeding before school. A brother with mechanical skills begins repairing pumps and boat engines. A cousin has a vehicle.
Another family joins them.
Perhaps five households pool their money for another cage.
Perhaps the fishermen’s association that already exists around wild capture fishing starts discussing aquaculture.
We started looking for an entrepreneur, but we found an entrepreneurial community.
And this matters because our imported model of entrepreneurship is incredibly individualistic.
We imagine the founder, the heroic entrepreneur, the person with the business plan and the equity.
It is essentially the Olaf model.
But where institutions are thin, communities frequently do internally what institutions elsewhere do externally.
They share labour.
They share information.
They distribute risk.
They create informal credit.
They find customers through relationships.
That can make capital accumulation more difficult, because successful people carry substantial social obligations.
But it can also make the spread of economic activity extraordinarily fast once the underlying constraint is removed.
There is a lot of entrepreneurial energy in the communities where African aquaculture is trying to develop.
We work with traders and smallscale farmers, and we have seen how they grow. They work extremely hard, extremely smart, and they are far more entrepreneurial than you realise.
The problem is that we keep presenting these entrepreneurs with an opportunity that requires them to build an entire industry before they can participate in it. Their time is better devoted to other activities with a higher chance of pay-off. Activities like trying to catch more fish, that might pay-off tonight.
So, why did salmon scale?
Norwegians did some extraordinary things.
The entrepreneurs were innovative.
The scientists were excellent.
Feed companies became world class.
Marketers eventually took Norwegian salmon around the world.
Government policy made mistakes too. The industry went through crises, overproduction, disease problems, bankruptcies and painful consolidation.
It wasn’t inevitable.
But salmon had something enormously important going for it:
An ecosystem was developing around the entrepreneur at the same time as the entrepreneur was developing the farm.
That is the lesson I think is worth taking to Africa: how do we recreate the economic conditions in which an African farmer can specialise as confidently as a Norwegian farmer eventually could?
That means commercial anchor farms.
It means genetics; feed; working capital; better roads; cold chain logistics; connected market traders; and affordable capex solutions.
It means government, business and communities working on different pieces of the same problem.
In short, it means accepting that we must build a farmed tilapia food system.
And perhaps that is why tilapia has taken longer than salmon.
We have been timing the farmer.
We should have been timing the industrialisation process.
Please note: Olaf and Manuel are fictional composite characters that I’ve constructed as a vignette to illustrate their different economic environments. The Norwegian historical elements draw on documented fisheries and aquaculture history; Manuel draws on the economic and operating environment around Lake Cahora Bassa and our experience of small-scale aquaculture in Mozambique.
