## Tuesday, January 26, 2010

### End of the recession?

The latest figures about to be released are supposed to signal at last the end of the recession. To recap a recession is defined as being at least two consecutive quarters showing a fall, but the end is defined as soon as any growth appears.

In other words you could have four quarters of decline followed by one small rise in the fifth quarter and hurrah the recession is over. The first problem with that is that markets are chaotic, one small bump in the graph doesn't mean it's time to celebrate, the second is dealing with percentages.

So let's start off at the initial value of 0%, for two quarters the GDP falls by 2% (therefore a recession) meaning that we're down 4%, for the third quarter it rises by 2% (marking the end of the recession) so we're only 2% down from where we started.

Now let's try that with figures. We'll start at 100, subtract 2% and we're down to 98, subtract 2% and we're down to 96.04, add 2% and we're back to 97.96 not 98. Now you can do this in another way by defining that initial 100 as the reference point so we start at 100 subtract 2% and reach 98 then subtract another 2% of the initial 100 making 96, then add 2% of the initial 100 and we're back to 98.

Note that I haven't changed the percentage figures in any way, but the results are different depending on how you calculate it. So anyone betting that all these graphs appearing in the media explain how they're calculating things?

Orphi said...

Maybe “the recession is over” will be a self-fulfilling prophecy, much like “there's a confidence crysis” was in the first place?

Aside from that, forget about trying to work out what official figures mean. This morning, I heard a report that some charty is shocked to discover than child poverty has increased dramatically in the UK over the last few years, while the government claims that their figures for the same period show it decreasing. Government conspiracies aside, how much do you want to bet they just measured things in a different way?

FlipC said...

Agreed, a bit like the stock market and the 'curb the bank' speech.

I did spot the child poverty story and the contradiction, but haven't had a chance to look at it further. Last time I did, if I recall correctly without looking, the official figures were x% below the median income, while the other was a set of questions about how often you go on holiday etc.

In that case not even a different interpretation of figures, but a completely different methodology.