Dropping import tariffs to raise exports: How mumbo-jumbo conquered the policy 'debate'

The heated discussion over the past few weeks about whether South Africa should unilaterally lower its import tariffs reminds me of Francis Wheen’s excellent book How Mumbo-Jumbo Conquered the World: A Short History of Modern Delusions. In it, Wheen documents a range of modern delusions where truths are assumed to be so obvious that there is no reference back to the underlying evidence, from the South Sea Bubble of 1720 to the fall of Enron.

So let’s go back to the original evidence: an academic paper — produced under the auspices of a National Treasury research project with Harvard University — entitled: South African Trade Policy Matters: Trade Performance and Trade Policy. The paper’s major claim is that trade liberalisation was one of the key drivers of non-commodity export performance since democracy and that further unilateral liberalisation is therefore appropriate in driving future export growth and diversification. These simple conclusions are packaged in a complex and apparently scientific set of econometric statistical techniques.

Let’s try to put the arguments in simple English and test how well they stack up. To use a crude analogy, saying that lowering import duties will raise exports is a bit like saying that by putting a flyweight in a boxing ring with a heavyweight, the contest will toughen up the flyweight until he is able to hold his own and maybe even beat the heavyweight. Theoretically lower import tariffs could raise exports by making domestic production less profitable relative to exports — but only if the domestic industry is already close to international competitiveness (the flyweight has become a mediumweight and is still packing on the pounds and muscle).

This is borne out by probably the most thoughtful piece I’ve seen written recently on trade policy by the United Nations Conference on Trade and Development (Unctad): Trade Liberalisation And Economic Reform In Developing Countries: Structural Change or Deindustrialisation?. The Unctad author shows that for the small group of countries that have achieved dynamic export growth, “economic reform, particularly trade liberalisation, has taken place gradually and selectively as part of a long-term industrial policy, after they had reached a certain level of industrialisation and development”. By contrast the remainder of developing countries in the study that “embarked on a process of rapid structural reform including uniform and across-the-board liberalisation” often experienced deindustrialisation and a reversal of their export patterns back to their resource-based competitive advantage.

In this light it is instructive to focus on the areas of non-commodity (mineral and agricultural) South African exports that grew relatively rapidly since 1994. The fastest-growing category of non-commodity exports is what the Harvard economists call “medium technology”. What exactly are these? Essentially they break down into two main groupings: capital-intensive resource-processing industries such as steel, chemicals and aluminium, and the automotive sector. So how much has trade liberalisation had to do with the export growth of these sectors?

The resource-processing industries have a long history of state support, many beginning their lives as state-owned enterprises such as Iscor and Sasol. These infant industries were nurtured by the state, privatised (albeit with no regulatory conditions attached) and their restructuring supported by generous tax incentives and restructuring arrangements in the early 1990s. The disappearance of sanctions meant that they no longer had to engage in secretive — and hence costly — channels for their exports but could sell freely into international markets. Now that these industries are internationally competitive, it is entirely appropriate to remove their import protection to lower costs in manufacturing and drive further efficiencies.

The automotive industry also has a long history of (not very effective) protection prior to democracy. The rapid growth in production and exponential growth in exports since 1995 is directly attributable to the Motor Industry Development Programme: a duty complementation programme that allowed exporters to earn offsetting import credits. This was one of the “sensitive” industries explicitly excluded from wide-scale liberalisation in South Africa’s tariff commitments that accompanied its re-entry to the World Trade Organisation in 1993. Import duties have come down substantially — from 80% in 1994 to 35% in 2006 — but in the context of an incentive structure that promoted rationalisation and export growth. This restructuring is not yet complete but requires building greater economies of scale and deepening of the automotive component supply chain going forward.

There are no silver bullets when it comes to the complex task of building export competitiveness and diversification. It involves dealing with a range of issues including removal of constraints related to financing, regulatory impediments, skills and technology development and infrastructure and logistics. What is clear is that an ideologically driven slashing of tariffs is no such silver bullet, unless we intend to aim it directly at our own heads.