Africa: growth outperforms

 As far as I know, no one else actually calculates quarterly GDP or monthly industrial production for Africa.   I've used higher frequency official data, where available, for industrial production and GDP, and annual data before that.  In some cases, I have estimated quarterly/monthly data using my variant of the Chow-Lin technique.  Ethiopia, for example, publishes no monthly or quarterly GDP or IP time series, so I have used annual GDP/IP time series and estimated quarterly series using the dollar value of imports as estimator.   My variant of the Chow-Lin technique forces the quarterly data to add up to the annual data (a kind of benchmarking), so the quarterly data should show similar rates of change and growth to the annual data, but they will give a clearer idea of cyclical turning points.

The countries included are the 10 largest (in terms of GDP) plus a couple of others, covering 78% of Africa's GDP.  Weighting is a problem, one I've been meaning to address for a while.  I have used PPP GDP for 2007, but I would like to switch to a system of variable weights and a chain-linked calculation, which will be much easier to do using constant price dollar-valued GDP (using the Maddison Project GDP estimates).   The variable weights calculation involves wrestling with the maths, the coding and the data, so I keep on postponing the evil day.  I'll get there eventually.  

As the first chart shows, GDP is less volatile than industrial production, which is what you'd expect.  It also suggests (since IP is more up to date) that GDP will likely go negative this year.  However, Africa's IP spiked up in May as Nigeria rebounded after a bizarre bank note crisis, but this spike has been removed by my extreme-adjustment algorithm.  So the decline in IP is not as fierce as it appears from the chart*.  Watch this space,

The second chart shows GDP for Africa relative to my calculation of world GDP.  (My world GDP calculation is a couple of quarters out of date --- I haven't updated it yet.)  African GDP has been outperforming world GDP since 2000, though before that growth in Africa was less than in the world as a whole.   Since the chart uses a log scale, a straight line would show a constant growth rate.  Note how, even though Africa continues to grow faster than the world as a whole, its growth rate slowed as world growth slowed after the GFC.  I'll talk more about this slowdown in world growth in later posts.

Africa makes up only a small percentage of world GDP (±2.5%), at least based on 2007 PPP weights, but its rapid growth means that that is more like 4% today, putting Africa as a whole as big as Germany.  This rapid growth is one of the reasons why I would like to move to calculations using moving weights.

The chart will be clearer if you click on it.
Note logarithmic scale.

In a sense, both the GDP and IP calculations are experimental, or at least, likely to be amended as I improve both the data and my programs.  However, the broad conclusions are unlikely to change.

* [Update:  see this post, which shows both the original and the extreme-adjusted year-on-year percentage changes in African industrial production]