One of the biggest barriers to deeper economic integration in Africa is the excessive trade costs that prevent regional trade. Import tariffs have traditionally been an important source of revenue for poorer countries, and it has taken several spaghetti-like agreements to reduce these. Although an agreement has been signed to create a Free Trade Area from the Cape to Cairo, none of the 26 countries have ratified it. Import duties remain between most African countries.
But tariffs are only of the costs of trade. It takes time to move a container from Johannesburg to Kinshasa, and the journey by land is often filled with tales of unscheduled delays and red tape. I remember traveling through the Victoria Falls border post between Zimbabwe and Zambia a few years ago and asking the truck drivers how long they had to wait to cross into Zambia. Their response: ‘A couple of days, if we are lucky’. This is no way to encourage regional trade.
Poor infrastructure is another significant barrier. The massive distances between major economic centres means that the unit cost of transport is high. A new paper in the Review of Economic Studies by Tufts University economist Adam Storeygard confirms this. Storeygard measures the impact of the oil price increases between 2002 and 2008 on the incomes of African cities. He compares two types of cities: those with a port on the coastline, and those of similar type but 500 kilometers inland. Using satellite imagery over the period, he finds that the oil price shocks increased the size of port cities by 7% more than in cities in the hinterland. The take-away: high transport costs retard growth. And because many African cities are located far from the coast, the high transport costs of poor transport infrastructure explains why African manufacturers find it difficult to compete with manufacturers in Asia and Europe. Just think of the difficulty manufacturers in landlocked countries like Malawi or Zambia face.
But even where better physical infrastructure reduces transport costs, other, ‘softer’ trade barriers often remain. Corruption, for example. Traveling into Malawi on my trip of a few years ago, we were pulled off the road a few kilometres after the border post by an armed man, and then required to return to the border post because we needed ‘additional insurance’. That was a $50 payment that went straight into the friend of the armed man’s pocket.
The effects of these ‘invisible’ trade barriers on trade and consequently economic performance have been hard to quantify, though, until now. In a new American Economic Review paper – ‘Corruption, Trade Costs, and Gains from Tariff Liberalization: Evidence from Southern Africa’ – Sandra Sequeira of the London School of Economics and Political Science finds that a reduction in tariffs between South Africa and Mozambique in 2006 had a very limited effect on trade. This is surprising: one would expect that lower tariffs would lead to higher levels of trade. And yet, the sharp decrease in tariffs had basically no effect (in technical terms, the elasticity of imports to tariff changes was very low).
What explains this surprising result? Sequeira uses a novel dataset of exporters’ bribe payments between South Africa and Mozambique to show that the decline in tariff rates at the border resulted in a 30% decline in the probability of bribe payments and a 20% decline in the average bribe amount paid. In other words, the lower tariffs did not actually reduce firms’ trade costs, it just shifted paying corrupt border officials to actually paying the tariffs as required by law, boosting government revenue. That is also why the elasticity of imports was so low: because costs did not fall in practice, there was no concomitant increase in trade.
Sequeira’s innovative study shows that high tariffs explain why corruption thrives. Remove the tariffs and the ability to solicit bribes vanishes. But don’t think that trade will suddenly blossom. Bribes keep trade costs lower than what they would be if tariffs were fully paid; lowering tariffs only lower the amount corrupt officials receive.
This has important implications for policy-makers: first, lower tariffs may actually result in an increase in tariff revenue as traders switch from paying bribes to paying the now more reasonable official tariffs. Free trade agreements (with zero tariffs) may not result in a significant fall in revenue either, because much of the revenue goes into the pockets of corrupt officials in any case, and will likely lead to greater transparency; Sequeira finds, for example, that trade statistics also improve when corruption practices decline.
But don’t expect free trade agreements like the one being discussed at the moment to result in a large increase in regional trade. As long as other barriers, like delays, severe red tape and poor infrastructure, remain, regional trade in Africa is likely to remain too weak to foster the economic development it promises to deliver.
*An edited version of this first appeared in Finweek magazine of 17 November.