In all the debates about free trade, we can forget sometimes that international trade is not optional for a very small, very poor country. If there are any kinds of returns to scale at all in many sectors, and casual observation and much research suggests there are, then a tiny domestic market will rule out any serious domestic production in many, many sectors (is the Gambia going to be making refrigerators any time soon?) So trade will be a necessity, specializing in what each small, poor country can do, and importing everything else. And what continent is full of small, poor countries? Africa, of course.
Then it’s all the more distressing that Africa does not have much in the way of trade links – shipping routes, air connections, Internet connections (for both communication and moving goods) – with the global economy. This is best seen in a series of pictures.
Lack of shipping routes going to Africa (except South Africa and Nigerian Oil):
Click here for a larger image.
Source: the excellent 2009 WDR “Reshaping Economic Geography” of the World Bank, Chapter 6
Scarcity of Airline Routes to Africa compared to rest of World (except South Africa):
Scarcity of Internet Connections except South Africa (map of # of IP addresses in 2007):
In short, Africa is disconnected from the global economy, which is very bad news for a continent that desperately needs international trade (the disconnection is both symptom and cause of the lack of trade). Lack of international trade = poverty for small economies.
This lack of trade links reflects many factors: rich country protectionism, domestic policies on customs & tariffs & foreign investment, poor port and road infrastructure, thus very high land transport costs within Africa, barriers crossing borders within Africa, Africa stereotypes that discourage foreign investors and out-sourcers, and so on.
There is some hope on the horizon for the Internet and communications issues, at least. High-speed fiber optic cables are connecting the east and west coasts of Africa to the rest of the world. And we all know about the famous cell phone revolution happening within Africa. Comparative advantage reflects your infrastructure quality and transport costs as well as your other endowments and experience. Perhaps the new broadband Internet connections will make possible all kinds of new businesses that economize on physical transport and use the Internet and cell phones instead. How about some kind of e-Africa.com, ready to open for business?
20 Comments
Interesting, but I don’t see much analysis here of *why* Africa’s trade links are bad. The answer surely has a lot to do with geography. In large part, Africa has poor sea links because more of its countries are land-locked, it has poor land links with Eurasia because of the Sahara, and it has bad internal transport infrastructure because it is comprised of poor, sparsely populated countries with difficult terrains and unforgiving climates. It’s all very well saying that things should be better, but we need to understand the valid reasons why they are not.
You said in the past that geography effectively doesn’t matter because African countries can just find their comparative advantage, whatever that is. But as you say here, the ability to gain from comparative advantage depends on infrastructure. Doesn’t it therefore also depend on geography? The World Development Report you cite supports this – it finds that “some divisions are beyond the control of individual countries”, and that if Switzerland was where the Central African Republic is its GDP would probably be some 9 per cent lower based on locational spillovers only.
African governments have other priorities than trade and infrastructure facilitating it.
A nit:
With respect to determining penetration / use of the Internet, counting IP addresses is not a very reliable metric.
The initial allocation of IP addresses was distorted based on a very limited view of the ultimate use of the Internet (e.g. certain universities and technology corporations received vast blocks of addresses). IPv6 was developed, in part, to deal with the problem of “running out” of IP addresses.
But what happened long before it was possible to adopt IPv6 was that a number of approaches were developed that tremendously reduced the need for IP addresses (principally, Network Address Translation, or NAT). This leads to a lot of distortion, and while many of the distortions cancel each other out, anyone familiar with this problem is going to have a hard time taking IP address distribution seriously as a development metric.
A good map of submarine cables tells a much more complete and nuanced story, e.g.: http://www.telegeography.com/products/map_cable/downloads/CableMap09_1600×1200.jpg
And no mention of submarine cables is complete without a pointer to my favorite article on submarine telegraphy: http://www.wired.com/wired/archive/4.12/ffglass_pr.html
Africa does need to better connect to the global economy. However governments, enterprises and other entities lack the capacities – e.g. information, policies, procedures, or infrastructure – to compete effectively in global markets and avail themselves of the advantages provided through international trade.
But efforts are being made to address these problems. In 2005 the WTO working with international partners launched the Aid-for-Trade Initiative to mobilize additional resources from the donor community and promote the importance of trade as an engine of growth in developing countries. The fundamental aim of the Aid-for-Trade Initiative is to help low income countries to overcome structural limitations and weak capacities that undermine their ability to produce, compete and maximise the benefits from trade and investment opportunities.
While the challenges remain significant, progress is being made. The OECD and WTO issued a joint monitoring report on aid for trade in July to put a spotlight on projects and programmes which aim to help African producers compete in the global market place. The monitoring of aid for trade is a work in progress and much remains to be done. The next stage will be focused on country-level dialogue, will be more evidence based and examine what objective evaluations can tell us.
http://www.oecd.org/document/56/0,3343,en_2649_34665_42835064_1_1_1_1,00.html
William Hynes
Policy Analyst
OECD/DCD/PRG
On my screen the second picture looks like a piece of abstract illustration. Pretty, but where did all those continents go?
In response to your internet hopes to Africa, I am sad to report than south africa’s broadband is slower than a carrier pigeon named Winston: http://news.bbc.co.uk/go/rss/-/2/hi/africa/8248056.stm
That is very interesting.
I had an unoriginal thought yesterday about the extent to which a lack of domestic and international demand for goods from Africa may substantially alter incentives there (especially the payoffs to education). Here’s my train of thought:
Assumption 1: most capital that embodies much technology is a substitute for unskilled labour, and a compliment to skilled labour. (Think of a combine harvester).
Assumption 2: The demand for this kind of capital is a function of the ratio of the prices of unskilled labour to skilled labour, the price of the capital, and the price and quantity demanded of the final good—that is, you will buy a combine harvester if drivers are plenty, peasants are rare, and people want to buy your barley.
Assumption 3: Due to the complimentarity between capital and skilled labour, for as long as there is little technologically-advanced capital, there will be few returns to getting skills.
Result: If the above three assumptions hold, and a country is in an equilibrium where there is little payoff to getting skills because there is no capital to use them with, and visa versa, then there exists an argument for international development bodies to act as “buyers” of final goods.
More buyers=higher quantities & prices=greater payoffs to investment and innovation=higher payoffs to skills.
Fantastic maps, by the way. I love maps.
A cursory exploration of African geography does not show why trade with the outside world is so small. A big reason is that Africa for much of its chaotic post-colonial history lacked four key drivers of economic growth – large urban areas to create mass demand for product, infrastructure to tie markets together, knowledge and skills centers to drive adoption and innovation and capital accumulation to stimulate domestic entrepreneurial activity.
South Africa provides the exception and is why it stands out as an economic power relative to the midgets the rest of the continent is. Those that could have provided an exception – Nigeria or Kenya for example have become failed development states as governance has become an avenue for looting than building.
But Africa’s fortunes may just be changing. Its fast urbanizing, getting better in governance and infrastructure slowly is being built at least in the telecoms space. Even ports have been privatized and are now taking advantage of the booming Africa-China trade. So Dr. Easterly lets look at Africa again from that perspective not from past development theory.
Frankly speaking, I don’t think Africa needs as many linkages to the world as it needs within itself. A lot of people here have posed geography as a factor inhibiting this absence of trade with the world. They have raised issues about countries being landlocked etc. In my opinion these are hardly problems in an integrated African economy. If we want Africa to trade well with the rest of the world, we have to strengthen regional linkages. Simply focusing on global trade linkages might will more likely create the same exploitative trade that has been going on over the past 400 years since slave trade.
Which came first, the chicken or the egg.Do you develop first and then get global trade links or vice versa!
I know it is now fashionable to beat up on Africa, but looking at the first map on shipping routes, Africa doesn’t look that different from South America.
It seems there is something about the cold wintry climates that favors shipping over the ocean.
With all the natural resources Africa has, it is in need of trade links because this would allow Africa to be self-sufficient at some point in time.
Africa need to be connected with the world.
The lack of infrastructure can explain the fact that none (except french and belgian multinational) want to invest in Africa.
Look a country like Cameroon or DRC; nothing have been done in those two countries to improve roads and other infrastructures facilities for decades.
How do you want to attract investors in your country if you are not able to provide electricity without ten interruption in a day? How could you attract tourist if there is only one flight a week to go to your country?
There is so many absurdities and nonsense in the way Africans leaders think.
One of the solutions to improve the situation would be:
1. oblige multinationals operating in Africa to publish the royalties that they pay to the African government and oblige especially the french multinational to pay the real price of what they extract.
the second one ; Promote regional economic bodies like ECOWA, EAC, SADC etc…. Only them could be enough powerful, if the African leader want, to conduct economic, financial reforms
When you get naked women off your mind you can really do some intelligent thinking. The nice thing would be to keep them off, at least in public and contribute. You will find in turn that they will appreciate you that much more.
SS
SS,
Your comment reminds me of that famous Robert Solow quote:
“Everything reminds Milton Friedman of the money supply. Everything reminds me of sex, but I try to keep it out of my papers.”
@ Salguod: An interesting point, and interesting map, too. On the map, it looks like Mogadishu is getting a broadband linkup? Is there an untapped market there?
The times are a-changing. East Africa is now benefiting from the SEACOM cable that came online in July. Four new cables are expected by 2011 in West Africa. The impacts could be huge. Good governance will be key. http://www.watradehub.com/index.php?option=com_content&task=view&id=1387
Excellent maps! Thanks.
I realized shortly after posting that bandwidth delivered to a country has obvious limitations as well. For instance the U.S. itself has very uneven delivery to actual people in many areas, despite setting at the crux of a very large fraction of global bandwidth.
Given that there is fairly decent geolocation available for IP addresses (the original map is from a company that does geolocation, in fact) what you want to do is measure something that is correlated to individual usage which originates from an African IP address. Even if measured from a US- or EU-centric perspective, it would tend to show use of the Internet for external-facing uses.
@Mike:
You can lead a horse to water, etc. If you read the Wired article I referenced above you’ll get a feel for the time lines that these things are planned on, and I suspect that at some point someone thought things were going to get better eventually. My guess is that the drop will be potential rather than actual until it makes some sort of sense.
@Iyinoluwa Aboyeji:
This leads to your point — if one looks at the submarine cable map of the U.S., one reason there can be such massive concentrations of cable landings in a few spots is that there is a tremendous amount of diverse and reliable internal bandwidth. No such luck in Africa (or S. America, for that matter). Just worked on a deal that dependent on a lot of intra S. America bandwidth and for a number of countries it was easier to run it through the U.S rather than from country to country. (This is only partly an infrastructure problem — it’s also a regulatory one, see below.)
@joe:
Governance indeed, and the willingness of countries to open the telecoms market to competition. My anecdotal experience in the the Caribbean taught me that bandwidth-dependent IT development wasn’t just a factor of landing the cable, but geting past the local C&W monopoly. Presumably there are similar problems in Africa.
@Fimbo
Perhaps you should also consider that the one thing Africans and South Americans have in common in that they are both conquered peoples-that naturally means the serve , sadly till today, their imposed “homelands”
@Douglas: Bandwidth in the US may be uneven but service provisioning is a lot easier in an America where 70-80% live in cities or near cities (and 60% on the eastern seaboard where most of the cables land) than an Africa where less than 40% do. Add the fact that America is at least 30 times richer than Africa and has the capital to lay nationwide fibre backbones unlike Africa which has to make do mostly with satellite (i.e. Nigeria, Ghana, Benin even after SAT-3).
Here are some marginally more useful statistics (although one wonders about numbers like the 102,000 Internet “users” in Somalia…)
http://www.itu.int/ITU-D/icteye/Reporting/ShowReportFrame.aspx?ReportName=/WTI/InformationTechnologyPublic&RP_intYear=2008&RP_intLanguageID=1
The methodology is here, and acknowledges the inherent flakiness of the “user” statistic — subscriptions are also apparently self-reported by the industry, which is likely prone to distortions as well.
http://www.itu.int/ITU-D/icteye/Indicators/WTI_Technotes.pdf