Localisation, the common ground in various government recovery strategies, will not lead to industrialisation, as it relies on reducing imports. Seeking to replace imports is no substitute for building the capabilities for exporting.

While doctorates will one day be written about the differences between the various economic recovery strategies published during the Covid-19 pandemic, one issue appears to have had significant common ground – localisation.

It is a word that appears in each of the economic recovery strategies published by the Presidency, Nedlac, the ANC, the SACP, the Tripartite Alliance and Business for SA. The only recovery strategy from a major “social partner” that does not use the word is by organised labour. But since it speaks of “local procurement”, a phrasing that covers similar ground, the absence ought not to be understood as labour not being part of the pro-localisation consensus.

The idea of localisation is an attractive one, and it emerges from a chain of reasoning. First, that SA is in desperate need of industrial jobs. Second, despite that need, a significant fraction of industrial inputs, intermediate goods and consumer products are imported from manufacturers beyond our borders, creating jobs in other countries. Third, we should therefore channel current demands back towards goods that are made here.

Since GDP is made up of the sum of consumption, investment and exports less the cost of imports, anything that reduces imports raises GDP if nothing else changes. In this way, localisation promotes growth, industrialisation and employment.

It is easy to see why this kind of reasoning is attractive to a range of interest groups. For importing businesses and their employees, it offers protection from foreign trade. For government, it offers an approach to industrial policy that is ideologically comfortable and which also reinforces other policy priorities including transformation and the creation of black industrialists. For all, it partly blames SA’s poor employment outcomes on low-wage producers in jurisdictions that are less worker-friendly and on our own fickle consumers. This is deemed to absolve government from responsibility for the country’s poor economic outcomes.

This report argues that the case for localisation is defective and unconvincing, that it is economically myopic, and it ignores the large but diffuse costs that must be paid by society even as it creates small but highly concentrated benefits for businesses whose output will expand.

We sought to understand the arguments for and against localisation as we understood them at the time. Subsequently, discussion has moved on, and commitment to the idea in government, organised labour and organised business has deepened. This report engages these substantial interventions.

What is localisation?

One challenge when discussing the pros and cons of localisation is that it can be difficult to be sure precisely what is being talked about as the term is used in different ways and can refer to different kinds of policy.

Sometimes, it is used narrowly to refer to the designation by government of certain specific products, which it will procure solely (or principally) from local suppliers. On other occasions, it is used to refer to a wider policy programme aimed at reducing imports across a broad range of product categories purchased by government, businesses and households.

Finally, it is also used to refer to policies to ensure that retailers stock the output of firms that happen to be located in their immediate vicinity and in the communities.

Products procured by government

Defined most narrowly, a policy of localisation refers to the use made by the Department of Trade, Industry and Competition (DTIC) of powers created in the Preferential Procurement Policy Framework Act 5 of 2000. These permit it to designate specific products that state departments must purchase from domestic manufacturers. To date, these powers have been used to designate 27 or 28 products for local procurement, though there have been repeated indications that this list will soon be expanded to 42 items, each of which will be championed by a senior figure from the private sector.

The products designated for local procurement are diverse, and it is hard to discern a set of guiding principles. Nor, indeed, does there appear to be one: in a 2020 presentation on the process of enforcing compliance with the list in government, the DTIC said that the basis for designation was “evidence indicating that the government buys product which is under distress caused by imports which displace local production and jobs”.

Localisation as import substitution

If the localisation of some of government procurement is one (narrow) conception of localisation, a broader one is captured in policy documents under the rubric of “industrialisation through localisation”. This is a barely modified articulation of an older approach to development called “import substitution industrialisation”, which many developing countries – including South Africa – embraced with varying success during the 20th century.

As with the narrower use of the term, in which localisation means that public institutions purchase a limited number of specifically designated products from local manufacturers, the idea of industrialisation through localisation presumes that demand for imported goods can be redirected to domestic firms, promoting expansions in output, employment and know-how in local manufacturing.

The current goal, in an agreement at Nedlac, is that 20% of non-petroleum imports should be replaced by locally manufactured goods within five years. According to the DTIC, “if executed successfully”, this would “return” over R200-billion in domestic demand to South African produced goods, and would raise GDP by over five percentage points relative to its present trajectory.

Similarities and differences

There are obvious similarities between narrow localisation, which focuses on public procurement choices, and industrialisation through localisation. The most critical is that demand (whether public or public and private) is redirected on the basis of policy-driven decisions relating to procurement choices, leading to expansions in output of local firms. To the extent that there is any consideration that there might be costs associated with implementing these policies, these are deemed temporary. This is, in our view, the critical issue.

Another similarity between these policies is that all are compatible with a range of policy goals associated with the transformation of the SA economy, not “just” its expansion and industrialisation. Thus, localisation policies can be shaped by and aligned with policies such as the DTIC’s “black industrialists” programme, the provision of support to black-owned and/or township-based businesses, and black empowerment.

It is possible to understand and conceive of black economic empowerment as itself a form of localisation policy in that it seeks to provide an advantage to local firms that happen to be owned by historically disadvantaged individuals. Notwithstanding these commonalities, it is important to understand that there are meaningful differences between narrow and broad localisation policies. The most critical is that the DTIC’s powers to compel the procurement of some kinds of goods exclusively from local providers are restricted to organs of state, and cannot be imposed on businesses nor compel compliance. Business’s reticence on this is peculiar, given that, in the absence of any plausible enforcement mechanism, it is inevitable that when the agreed target of replacing 20% of non-petroleum imports with local products is missed, it will be blamed for its failure to deliver.

What’s so bad about localisation?

The principal problem with any version of localisation policy is the assumption that it is something of a free lunch: demand is transferred from imported to locally produced products to benefit local firms, its employees (actual and prospective) and owners. To the extent that there are positive externalities associated with local production, this will also accrue straightforwardly to the national economy. But the supposed costlessness of the switch from imports to local products demands some interrogation.

Costs of localisation

As noted above, the DTIC has already used its powers to designate 27 products that government agencies must procure solely or principally from local suppliers. What is wrong with that?

“When products are designated for local procurement,” Professor David Kaplan pointed out, “it means that those products were not the buyer’s first choice because, if they had been the first choice, there would be no need for the policy. But if the local products are not the first choice, then there must be a reason for that. Are they more expensive? Are they inferior in some way? There must be a cost to choosing the local product over the imported product, otherwise there would be no point to the localisation policy.”

The structure of localisation policy gives local suppliers the power to raise prices. Professor Lawrence Edwards said: “What you are talking about when you talk about localisation, is a quantity restriction: you are saying that some pre-determined percentage of a particular product has to be procured from local suppliers. This is very inefficient because if you are going to fulfil the requirement set by the policy, you are going to have to buy it from a limited number of local sources. Inevitably, those sources can then charge you more than they would have been able to if there was foreign competition.”

The key problem here is that the policy provides support to local producers by imposing a quota on government, which must buy some percentage of a particular product from local firms. This quantity restriction creates no upper boundary on the price local firms can set. There are other costs to consider, too. Costs over time can actually result in a decline in local firms’ competitiveness.

Reducing firms’ efficiency incentives

To the extent that localisation works to shift demand, it means that consumers – in this case, government – must pay higher prices or accept a product that is of inferior quality or inappropriate. But the effect is dynamic: by divorcing the domestic firm from competitive pressures, the localisation policy reduces the incentive for domestic firms to improve efficiency and the quality of their products.

“The very large downside risk,” Edwards said, “is that you end up supporting very inefficient firms. This is especially true in South Africa where industries tend to be highly concentrated, so there is limited competition between local firms to supply products that government agencies are not allowed to purchase elsewhere.” Worse still, because prices for goods are inflated, firms may choose to increase their margins by lowering the quality of their product.

Designating some products as having to be purchased from local manufacturers means paying a higher price for those products or obtaining goods of an inferior quality. It also creates the distinct possibility that the price and quality differential between local products and imports will grow because local firms are no longer subjected to the discipline of foreign competition, while any innovations and quality improvements made by foreign manufacturers become inaccessible to government.

Nor, finally, is that the last of the malign effects of localisation policies on firms’ incentives to increase efficiency. Firms, knowing that government is amenable to arguments about the benefits of localisation, will invest some of their time, energy and resources in lobbying government for designation rather than in increasing their output.


Perhaps the most important thing to say about localisation policy (in its variants) is that SA’s poor economic performance over the past 13 years has nothing to do with any supposed “overpropensity” to import, and everything to do with a range of self-inflicted policy and governance injuries.

These include our inability to keep the lights on, our decaying infrastructure, the fact that the container terminals at our ports are both very expensive and inefficient, the deepening fiscal crisis, rising levels of lawlessness, and our education systems’ inability to produce adequate numbers of skilled workers.

Any policies that seek to strengthen local industry are likely to appeal to a range of stakeholders. Nevertheless, the claims made on localisation’s behalf are simply not credible. It is important to recognise that the DTIC has sought to justify the localisation policy by a focus on the occasional success story, and its approach will be loudly supported by the firms that benefit. The broader costs of the policy, borne by government, SA consumers and by exporters, will be ignored.

One of the ironies of this is that localisation, precisely because it raises costs, runs in contradiction to the President’s repeatedly stated commitment to bring down the costs of doing business and the national goal.

Protection, import substitution and industrialisation through localisation all have very spotty records as vehicles for promoting development. By contrast, exports have been integral to successful development.

South Africa’s future growth will similarly be heavily dependent on growing exports. Seeking to replace imports is no substitute for building the capabilities for exporting. An undue focus on localisation and on prohibiting imports will weaken our export capabilities. Localisation should be called for what it is: an anti-export strategy, one that will only further constrain our future development.

This is an abridged version of the full CDE report “The Siren Song of Localisation: Why localisation policy will not lead to industrialisation” based on discussions with Professor Lawrence Edwards and Professor Emeritus David Kaplan, both from the School of Economics at the University of Cape Town.

Article published by Daily Maverick