Head-in-the-sand industrial policy

Learning from the great Trans-Saharan Ostrich Expedition

In 1910, the British consul in Tripoli sent a feather to Cape Town. The plume, writes Sarah Abrevaya Stein in Plumes: Ostrich Feathers, Jews, and a Lost World of Global Commerce, ‘was of medium length, fairly broad, with a remarkably dense, strong quill’. Its origins in North Africa were mysterious, but the consul noted that it arrived with a party of Arabs ‘who had crossed the Sahara desert from Southern Soudan’.

The feather caused a stir. For months, Cape bureaucrats had been diligently collecting feathers from wild ostriches across Africa, aiming to procure the most coveted plume: the elusive Barbary feather. All previous attempts had been disappointing; the Moroccan specimen was notably described as ‘long, narrow, drooping’. Yet, the Tripoli feather was different. It precisely matched the tastes of buyers in Europe and America. If the bird that produced such feathers could be found, brought to the Cape, and crossbred with the hardy local ostriches, it could produce plumes capable of dominating the international feather market.

Thus began a secret, government-sponsored mission: the great Trans-Saharan Ostrich Expedition. (This should 100% be made into a movie.) Stein describes how, led by Russel Thornton, an agricultural expert of the Cape Colony government, the expedition set sail in August 1911. Thornton, accompanied by Frank Smith and JMP Bowker (‘competent authorities on the subject’), travelled via London and Paris, docking at Forcados at the mouth of the Niger River. From there, they journeyed by paddleboat to Baro, hired 107 Hausa porters, and moved onward – by horseback, foot, and train – to the camel caravan entrepôt of Kano. Once in Kano, the team painstakingly checked every feather caravan that arrived. The work was arduous, the sandstorms punishing, and the caravan leaders protective of their wares.

Smith records their eventual success in his unpublished memoirs:

After many disappointments a parcel of feathers of undoubtedly the right type was brough in one day for sale. These were immediately pounced on and the vendors eagerly questioned as to where the feathers had come from. They turned out to be a party of Tuareg Arabs who had come from a district beyond Timbuktu. They stated that their particular district was isolated by a strip of hundred miles of desert from any other habitable part. They also said there were numerous ostriches there, both wild and in semi domestication, and that it lay in the French territory.

The expedition immediately cabled the happy news to Cape Town. Would the government approve funds to secure this treasure? Yes, came the response: £7,000 was allocated for the procurement of 150 ostriches. (At the time, £7,000 was roughly equivalent to 1,000 sheep at the Cape – or approximately R3 million in today’s terms, using an average price of R3,000 per sheep.)

Before the birds could be acquired, however, they needed permission from the French authorities. The French commander promptly refused. Unwilling to accept defeat, the team pushed forward into French Sudan, pursued – or so they believed – by French and American spies, the latter allegedly sponsored by an ambitious American feather magnate. To evade capture, the expedition split into three separate groups, later reconvening in Timbuktu. There, with assistance from the emir of Katsina, they finally secured 156 prized ostriches.

The next challenge, perhaps the most daunting of all, was bringing the birds safely back to the Cape. Remarkably, when the shipment arrived in the summer of 1912, 140 ostriches had survived the grueling journey. Men and birds alike were welcomed with great fanfare. Thanks to the government’s bold intervention, the future of the Cape ostrich feather industry seemed secure.

Only, it wasn’t. Before the imported birds could be crossbred, the global feather industry collapsed. Henry Ford’s automobile had made extravagant feathered hats – popularised more than a century earlier by Marie Antoinette – impractical. Demand for feathers of any kind evaporated almost overnight. Then came a global war. Luxurious plumes that once fetched astronomical prices were now fit only for feather dusters. In Oudtshoorn, the global heart of the ostrich feather trade, fortunes vanished as rapidly as they had appeared.

Industrial policy is once again back in fashion. Simply put, industrial policy involves direct government intervention to foster growth in specific economic sectors, typically through subsidies, tax incentives, or targeted regulations. Two years ago, I outlined how prominent economists, notably Harvard’s Dani Rodrik, have persuasively defended industrial policy, citing positive externalities, coordination failures, and activity-specific public inputs as clear theoretical justifications for government intervention. Yet despite these compelling arguments, practical implementation remains fraught with problems. Politicians and bureaucrats inevitably choose winners and losers, decisions that are vulnerable to capture by special interests or political expediency. In South Africa, this risk is especially pronounced. I previously argued that our industrial policies prioritise redistribution and political convenience at the expense of genuinely growth-enhancing goals such as export promotion and technological innovation.

What I didn't fully appreciate then was the speed at which technological change can render even the best-designed policies obsolete. Recent research by Réka Juhász and Nathan Lane highlights how modern industrial policies now lean heavily on subsidies and export incentives rather than traditional tariffs, Trump’s tariffs aside, of course. Their crucial point is that even carefully targeted measures can quickly become ineffective when technological circumstances shift unexpectedly. The global semiconductor sector provides a vivid example. Studying this industry, Pinelopi Goldberg and colleagues demonstrate how cross-border technological spillovers can swiftly undermine any competitive advantage that subsidies initially confer. The reality is that industrial policies are always implemented under conditions of profound uncertainty.

South Africa is currently drafting a new industrial policy. Minister Parks Tau declared in July last year that ‘the industrial policy will be the centrepiece of our economic development strategy, mobilising an all-of-government and all-of-society approach’. I have no idea what that means, but it does raise many questions.

My first concern is that South Africa doesn’t have one plan – it has many. The dtic has already produced eight ‘master plans’ spanning sectors as varied as poultry, textiles, medical devices, and global business services (call centres). How these individual plans will integrate with a broader industrial policy remains unclear. Moreover, there are many other policies that overlap with industrial policy: labour policy, infrastructure plans, tax rules, small business development strategies, and myriad other documents, all the responsibility of different departments.

A second concern relates to what exactly we mean by ‘industrial policy’. Are we restricting our attention to traditional manufacturing, or should we also embrace the vast and heterogeneous services sector? The inclusion of call centres in the dtic's master plans suggests a broader definition. But what exactly would an industrial policy look like for such a diverse industry?

Thirdly, there remains considerable uncertainty about who will shape the new policy. Will it primarily be the dtic, or will Ramaphosa’s economic advisory council, Operation Vulindlela, the National Development Plan, NEDLAC, BUSA, IDC, or Treasury also play significant roles? Each of these entities represents different interests, even contrasting ideologies. Some may view industrial policy primarily as a mechanism to boost economic growth, while others may prioritise racial transformation or redistribution. With such divergent objectives in play, the prospects for a coherent, clearly defined industrial strategy seem remote.

Finally, and crucially, there is the ever-present problem of uncertainty. Even the best-crafted industrial policy from three years ago could not have anticipated the disruptions caused by artificial intelligence (in technology) or Trump (in politics). We inhabit a profoundly uncertain world where agility and flexibility have become the most valuable traits for any organisation, government included. An overly inclusive policy document, diluted by compromises and conflicting interests, risks being meaningless at best and actively harmful at worst.

A century ago, the Cape government believed it had found the perfect industrial policy: a bold plan to dominate the global feather market. They did everything right, mobilising government resources and taking significant risks. Yet unforeseen technological advances and global politics rendered their careful planning futile. As South Africa embarks on yet another ambitious industrial policy, we must heed history’s warning. Good intentions alone offer no insurance against an uncertain future. Let’s not bury our heads in the sand to yesterday’s lessons.

An edited version of this article was published on News24. Support more such writing by signing up for a paid subscription. The image was created with Midjourney v7.