Become complex or stay poor

African rankings
All this has important conclusions for African economies, which have long sought to diversify their economies, partly to make them less susceptible to the cycles of commodity prices, but also to make them richer. The lesson that the Atlas is trying to teach here is that diversification is not enough: to get consistently richer, economies also need to develop industries that are more complex, that can make things that fewer other countries can make, and because of this, command higher prices and create bigger profits.

If that argument is correct, the biggest question for policy makers is ‘How do we get from here to there?’. Unsurprisingly, it turns out that moving from a simple economy to a complex one is not easy. The worst-case scenario is an economy where there is high dependence on one very simple industry that has very few connections with other industries. Nigeria is a case in point. Sudan is another. Both have significant oil revenues; neither has anything else in the economy that can match oil in terms of attracting investment and developing complexity. The oil revenues themselves provide a disincentive for diversifying up the complexity scale.

For Africa, the bad news is that an alarming number of countries are at the bottom of the complexity scale. The Atlas does not include all countries – it is limited to countries for which there are adequate trade data, where there is a population above 1.2m, where exports are at least $1bn a year between 2006 and 2010, and where data is reasonably reliable.

That leaves 128 countries, including most of Africa. Of those 128 countries, no African economy makes it into the top 50. According to the rankings of complexity in the Atlas, the most complex sub-Saharan African economy is – no big surprise – South Africa at number 55 (and in the whole of the continent, only Tunisia is ahead at number 47).

Then comes a group of economies that are still relatively complex in African terms – Namibia, Kenya and Senegal, ranked at numbers 72 to 74, followed closely by Mauritius (77), Zimbabwe (80) and Uganda (87).

The really bad news is that in the bottom 10 countries in terms of economic complexity, there are eight sub-Saharan countries, plus Libya. Mauritania is ranked as the least-complex economy in the entire world ranking, with Sudan at second-last place and Angola third last. Also in the bottom eight countries are Congo, Guinea, Cameroon, Nigeria and Gabon (DRC was excluded from the Atlas due to shortage of data).

Significantly, all of this last group are largely dependent on oil (with the exception of Mauritania which exports some oil, but is mainly dependent on unprocessed metal ore exports). Thanks to oil, most of these countries are not poor in GDP per capita terms, but according to the Atlas they are very poor in terms of ability to grow in any way other than growing by growing oil revenues – and of course, oil eventually runs out.

The biggest question all this leaves for policy makers is this: if complexity is important for long-term growth, how do you get from a simple economy to a complex one? This is the problem that all developing economies must wrestle with – the fact that industries tend to develop and grow into similar industries.

According to Professor Hausmann, the problem looks like this: if you have a hospital with no operating theatre, no surgeons and no anaesthetists, and you have limited finance so you can’t just go out and buy all three, you face the challenge of complexity. There is no point in having an operating theatre without anyone to operate in it. There is no point in recruiting an anaesthetist if there is no surgeon and no operating theatre. In short, it is very difficult to engender complexity.

Economies have to develop by moving into areas of activity that have some similarity to what they are already doing. If you have industries that show a lot of connectedness to other industries, your growth outlook is relatively good. And the more complex those new industries are, the better able you will be to manufacture the more profitable goods that others find it difficult to make. If your industries are the kind of outliers that don’t connect well with other industries in terms of the skill sets required, then you face a big problem. You are stuck where you are.

The Atlas ranks industries in just these terms: how connected they are. It does this rather simply – instead of looking into every industry and calculating what skills and technologies it requires, it simply calculates the likelihood based on the export data of a country exporting all possible pairs of products.

If it is more likely that a country exports shirts and tablecloths than shirts and jet engines, then shirts and tablecloths are closely connected and shirts and jet engines are not. Using this calculation, the Atlas gives a view of the future, called the Complexity Outlook Index. The Outlook tells us how likely it is for new, connected industries to develop.

Interestingly, when it comes to the ranking of the complexity of products rather than their connectedness, it turns out that industries that produce very hi-tech products such as computer chips are not the most complex in terms of the skills required to manufacture them. Of the top five most complex products, two are machinery designed for metal fabrication, and the other three are chemicals and healthcare products – chemical analysis devices, X-ray machinery and photographic chemicals. The five least-complex products are all unprocessed or part-processed agricultural produce: raw cotton, tin ore, natural latex, sesame seeds and cocoa beans.