A tweet from @inquestioner went “Scary,34 million joined the ranks of the poor in India because of the recession that everyone was in denial of.UNDESA figurs.” [sic]
I usually find these figures very global-warming type – meaning, highly suspect and jugaad methodology - so I thought I’d investigate. In the “World Economic Situation and Prospects, 2010” report, Page 35, they say:
The reduction in employment and income opportunities [due to the slowdown] has led to a considerable slowdown in the progress towards poverty reduction and the fight against hunger. Estimates by the Department of Economic and Social Affairs of the United Nations (UN/
DESA) suggest that, in 2009, between 47 and 84 million more people have remained poor or will have fallen into poverty in developing countries and economies in transition than would have been the case had pre-crisis growth continued its course (table I.3). This setback was felt predominantly in East and South Asia, where between 29 and 63
million people were likely affected, of whom about two thirds were in India.
So the figures for Asia are between 29 and 63 million people – that is a huge enough range for me to say “WTF” anyhow, but let’s not digress – and 2/3rd are in India, so for us it’s between 20 and 42 million people. 34m is somewhat in the middle if you are slightly mathematically challenged, but let’s not bicker about a few million here and there.
The important thing is how they arrived at it. They said – okay, India’s growing at 9%. If it continued to grow at 9%, then X number of people would be poor. But because of this slowdown, we have Y poor people, and Y is larger than X. Therefore, this figure Y-X is the number of people that have been reduced to poverty by the slowdown.
Let’s not talk X and Ys. Let’s talk real numbers.
Assume we had a 100 poor people. Let’s say that for every 1% growth, we reduced poverty by 1 person. So with 9% growth, we are left with 91 poor.
But we grew only 6%. So, post the recession, we have 94 poor people. 94 is less than 100. That might sound like a good thing. But no.
The UN-DESA way of looking at this – and their glasses must forever be half empty – is: Goddamn recession did not take 3 people out of poverty, so 3 people got poorer.
Another way of looking at it is: 6 people got out of poverty. 6 is good. Even 1 is good, come to think of it, but 6 is definitely good. And it is wrong to focus on the 3 number.
For one, 9% growth was not sustainable; it was extraordinary. It was likely to be 6-7% averaged over years, so one year we’d do more, another we’d do less. Extrapolating a pre-crisis growth figure is silly; in that way, I could extrapolate the “Hindu Rate of Growth” of 3% pre-1992 and say damn, 600 million people are out of poverty today since 1992 because we grew more than expected.
Second, the focus on the smaller figure throws real achievement out the window. There are no real figures in that sheet – but I can imagine that if this calculation yields 34 million “poorer”, then the absolute number of people we got out of poverty must then exceed 60 million. That means, we got 6 crore people out of poverty, post recession – less than the 9 crore we expected. That statement paints an entirely different picture.
Finally, I must say the statistics are screwed up because it uses per-capita income to determine poverty levels – yet, if money was earlier more concentrated in a few rich people, and post recession got better distributed among the entire population, the poverty figures UN-DESA quotes will be overstated. They acknowledge this, though:
It should be noted that the estimates presented here take into consideration the impact of the downturn only on growth in income per capita compared with continued pre-crisis trends. Hence, these should be interpreted in the first instance as a slowdown in poverty reduction owing to a drop in the mean per capita income of developing countries. For lack of additional information, the estimates do not take into account likely changes in income distribution.
That statistic is simply gleaned from macro-figures. We aren’t told how many people are really poor (buying power wise), how many of these are urban/rural, how many are in organized/unorganized sectors, what’s the birth/death impact and how the numbers are moving. Those stats may tell us where we need to focus, where achievements are good and where we can improve. What we can’t afford is to have global statements like “the recession made 34 million people poorer” – because it is a hollow statement with conveniently extrapolated matter, and yields nothing. It truly makes us poorer.
Girish Shahane has a good post about this too.
So once in a while, I will rant. This is that time. Thanks for listening.
P.S. Why is this relevant to trading? One comes across this kind of fallacy all the time – the subprime crisis was exacerbated by people running excel sheets extrapolating data and lost connection with logic (once it boils down to “House prices always go up” the logic failure is evident). LTCM failed because they assumed “Normal distributions” but it didn’t live up to its name. And the efficient market theory fell flat on its face because despite all sorts of mathematical proof, it simply was non-existent. (My feeling is that it’s attractive to a lot of people for it’s simplicity, so they like indexing/asset-allocation/passive investing even if the the data may not apply to India)
