Email to Thomas Pogge.  Appears to be reply to email of 21 June 2003.


RE: Economics and poverty statistics	

Dear Thomas  

Thanks.   Kanbur's paper is at http://www.arts.cornell.edu/poverty/kanbur/Pov&Death.pdf  .

My own analysis of the implications of differential mortality for welfare economics is more comprehensive.     

The paper "1015" (www.mattberkley.com/1015.htm ) is an economics paper.    It shows some of the flaws involving the use of population averages, and measures of income inequality, as well as drawing together a number of issues and making a proposal to the UN.   

The paper on utilitarianism (www.mattberkley.com/utility ) is a philosophy paper.    It clarifies the ambiguity in the use of the word "utility", often applied by economists to people's welfare level at different times.     That isn't a Benthamite concept.    

I'm also developing a set of guidelines for the UN's research on poverty.     The requirements are so low that people would look silly if they failed to sign up to them.      For instance, economists and statisticians obviously should only use income or expenditure data with information on the inflation rate for the poor.      Simple really, but I think it's possible to rein in economists' enthusiasm for coming to conclusions which their data don't support.     

The paper "1015" mentions data quality issues.      One of the features of research on world "income poverty" is that the main dataset is known to be unreliable.     Put that together with the inflation flaw and the mortality flaw and we can see the weak evidential basis for all economists' statements about large-scale "poverty" trends.      The whole debate about "growth" and "distribution" relies on income without data on prices, as do all other debates about the effects of policies on poor people.   

I argue for life length statistics as more practical indicators for hungry people than the vastly complex task of using income data.     Again, the paper "1015" makes (in this case fifteen) brief points.   

Best wishes

Matt