On 21-Dec-2011 John Bibby wrote:
> I half-heard a discussion on the Today programme this week when child
> poverty was being discussed. John Humphrey derided the current measure
> which is based on 60% of the median income i.e. 60% of the 50th percentile.
> (I'll call this the 60/50 point) He said that in a time of recession, the
> median would drop, and the number of households below the 60:50 point could
> drop. So paradoxically, although we are all getting worse off, the 60/50
> measure would indicate fewer households in poverty. (So presumably would
> any p/q measure, for all p and q.)
>
> I think he has a point here - but what exactly is the point? And has
> anybody analysed this in a systematic way? (Clearly for any income
> distribution, the poverty-value is unchanged by any linear transformation,
> for all p and q - but what about non-linear transforms? And how to model a
> changing population??
>
> Can anybody throw any light on this please?
>
> JOHN BIBBY
>
> PS: Alan Milburn was involved too - see
> http://www.guardian.co.uk/tv-and-radio/2011/dec/18/
> today-child-poverty-alan-milburn
That was bad (statistical) logic on John Humphreys's part
(and also on the part of the quote in the Guardian article).
60% of the median income is at a quantile which depends on
the distribution of income. When incomes change, as a result
of recession, the distribution may change. The median may
fall, but perhaps most of the people previously between
60% and 100% of the previous median may drop below 60%
of the new median (the people previously below 60% also
remaining below 60% of the new median). Then the proportion
of the population in poverty would rise dramatically.
Or it could of course cut the other way (but not for the
reason John Humphreys suggested).
One big objection I have to the "60% of the median" definition
is that ensures that at most 50% of the population are in
poverty. One could imagine a society in which the few oligarchs
of corporate business have pocketed so much of the national
wealth that nearly the whole population was objectively in
poverty (i.e. unable to afford adequate food, let alone
pay their rent or their mortgage). The official definition
completely evades defining "poverty" in terms of "not being
able to afford the nasic neccessities of life".
For similar implications of the "60% of the median"
definition see my posting to RadStats of 07/06/2010:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind1006&
L=radstats&P=R2147&1=radstats&9=A&J=on&K=3&
d=No+Match%3BMatch%3BMatches&z=4
[= http://tinyurl.com/czf6z5u ]
which concludes:
Take all those with disposable incomes above 60% and
up to 100% of the median. Reduce all their incomes to
below the current 60% level. Then, suddenly, many of
those who had been below the poverty line will be above it,
with respect to 60% of the new median. With care, none
of those whose incomes have been lowered will subsequently
be below the new poverty line. Thus many will have been
removed from poverty, none will have been brought into
poverty, and a profit will have been made. Magic!
Best wishes, and Season's Greetings (according to your
preferred terms) to all,
Ted.
----------------------------------
E-Mail: (Ted Harding) <[log in to unmask]>
Date: 21-Dec-2011
Time: 17:18:53
This message was sent by XFMail
----------------------------------
******************************************************
Please note that if you press the 'Reply' button your
message will go only to the sender of this message.
If you want to reply to the whole list, use your mailer's
'Reply-to-All' button to send your message automatically
to [log in to unmask]
Disclaimer: The messages sent to this list are the views of the sender and cannot be assumed to be representative of the range of views held by subscribers to the Radical Statistics Group. To find out more about Radical Statistics and its aims and activities and read current and past issues of our newsletter you are invited to visit our web site www.radstats.org.uk.
*******************************************************
|