From: Matt Berkley@aol.com
Sent: 10 October 2001 01:00
To: sudhir.anand@economics.ox.ac.uk
Subject: Statistics and survival
Dear Professor Anand

Thank you for listening to me the other day.

I have sent a letter to the Economist, which I attach here.   

I believe that I am raising a fundamental issue for social science about the validity of using statistical averages to infer  benefit to individuals, in populations with variable mortality.     

Essentially, I am saying nothing more than Rowntree said 100 years ago  -   the death rate is the best instrument for measuring the physical well-being of a population.   He found that the death rate among the poorest was twice the rate among the best-paid of the working class, but this is exactly the kind of statistic that I cannot find on poor countries.  

I have discussed the Dollar and Kraay paper on growth with Frances Stewart, Meghnad Desai, James Putzel, and others.   Considering their comments, it is surprising to me that it is accepted by DFID.    

I have discussed differential mortality with a DFID statistician, the head of statistics at the OECD and the head of statistics at the World Bank.   None of them can provide any data on possible effects on average income statistics, or on poverty reduction.   

I have many more points about this than I have made in my letter.     

If you would like to discuss this further, my number is Oxford 727092.

Yours,


Matt Berkley