This economy of ours!

Economic data out for South Africa is painting a good, if slghtly confusing picture. The good news is that the economy is still growing healthily (not compared with China’s official growth though) and inflation has dipped down very slightly. Neither of these are truly unexpected though. Exchange rate appreciation and a relaxation in the dollar oild price will have had a muting impact on inflation.

CPIX (the basket excluding things like mortgage repayments and a few others) increased by 5% year-on-year in October. We only have data for October available now because of delays in collecting, collating and analysing data. This was slightly above forecasts, but nobody’s panicking yet. The same figure for September was 5.1%.
Meanwhile, growth in GDP is also continuing (4.7% annualised) in the third quarter, without showing much impact of the interest rate increases Tito Mboweni has put in place this year. Having said that, if one digs down into the sector details, the interest rates increases have not been completely ignored with property-related sectors showing markedly reduced growth from earlier levels.
Now for the confusing part. The previous quarter’s GDP growth has been upped to 5.5% from 4.9%.  First quarter figurees have been upped substantially from 4% to 5%, but last year’s growth adjusted only slightly from 4.9% to 5.1%.

So why all the changes? Well, before everyone goes on about “Lies, Damined Lies, and Statistics”, one should understand that estimating GDP is a tricky task in a developed economy, let alone one with a signficant contribution from an informal sector with limited records and reporting. As it is, there are discrepancies between information such as VAT receipts, money flowing through banks and the official GDP figures. While these measurements will be affected differently by different things, they should have a strong relationship to each other.

I’m going to dig into this as well over the next few months, but any comments or inputs are very welcome!

Urban legends and cocktail conversation

I’ve had an idea to analyse some of the many topics that come up in conversation time and time again. Chances of winning the lotto, FNB’s Million a Month account, randomness of the iPod shuffle to name a few. I’ll try to get hold of some interesting datasets and perform some basic analysis. My aim?

  1. To find out whether any of the often-claimed techniques might actually work
  2. And so discover what sort of randomness is hidden within these events
  3. And also see which of these datasets I can easily get my hands on. Any thoughts on my chances of getting FNB to chat to me about how they select winners?

Anybody have some data they’d like me to look at? Any more questions to add to this list? I’ve been away from blogging for a while, so let me know when your office Christmas party is so I can prepare your answers in time!