Our Word is Our Weapon

In “Trade Liberalization and Poverty Reduction in Sub-Saharan Africa“, Marian Tupy says

[Sub-Saharan Africa] continues to be one of the most protectionist regions in the world … Whereas average applied tariffs in high-income OECD countries fell from 23.7 percent to 3.9 percent between 1983 and 2003 (a reduction of 84 percent), average applied tariffs in SSA fell only from 22.1 percent to 17.7 percent (a reduction of 20 percent).

His source is this table from the World Bank. Scanning it, something struck me as odd: why did Tupy take 1983 as his base year, when there were only three Sub Saharan African countries reporting figures for that year, compared to nine in 1981 (the first year available)? There seems to be no good reason … but wait! If you take the average of those nine in 1981 you get an average applied tariff rate of 32.2%, much higher than the 22.1% in 1983 - and that means a much bigger reduction in trade protection to 2003. I make it about a 47% reduction, which seems to be in line with the world average.

That conclusion, of course, would blow a hole in Tupy’s argument. So it looks like he just ignored it and plumped for the more unrepresentative figures (3 countries standing in for a whole world region, remember) that did lend some credence to his case. Good to see standards are as high as ever at the Cato Institute.

Update: Jonathan Dingle points out another fanciful finding from Tupy: one of his figures “makes it appear that Sub-Saharan Africa’s trade regime was the most liberal in the world in 1983″.