From 2007:

The "dollar a day" and "two dollars a day" statements from the World Bank are based on what are known as PPP dollars.

PPP stands for "purchasing power parity", but the system has not considered specific prices relevant to the "poor".

Things are generally cheaper in "poor" countries than in the USA. So the World Bank dollars are in those countries generally worth a fraction of a real dollar.

But what people on one or two World Bank dollars a day can actually buy has not been estimated by the World Bank.

This problem is compounded by the mathematics of price indices. A related problem concerning representativeness affects measures of prices and inflation in general.

National price indices, and therefore inflation indices, are more influenced by luxury items than by basic items. That follows logically from a mathematical truth: the total money spent on a category (prices multiplied by sales) determines that category's influence on the index.

This means that price and inflation indices are mathematically biased in favour of the "rich" - if by that we mean people who buy expensive things and/or many things.

Suppose prices of cheap items bought by the "poor" go down. Suppose that is outweighed by rises - smaller in percentage terms - for expensive items of the "rich". Then a macroeconomist might say "the poor have not got richer" when in reality they could now buy more with their money.

Similar considerations apply in relation to country comparisons.

Since the World Bank system did not look at prices for the "poor", it is perhaps difficult to see the term "purchasing power parity", in the present context of poverty, as anything other than a misnomer.

The question of the appropriateness of the word "parity" despite the lack of systematic purchasing power estimates for the "poor" (and thus the impossibility of basing any judgement about "parity" of purchasing power of the "poor" between countries) is in addition to the general problem:

As a result of economists' and politicians' use of the word "dollar a day", many people may falsely believe that those on a "dollar a day" actually spend, in local terms, two, three or four times what they really can.

Matt Berkley 17:15, 17 March 2007 (UTC)