I don't know anything myself about the way the PPP is constructed or the neoclassical assumptions that Paul proposed were used. Intuitively, though, it makes real sense to select the PPP measure (ie., something that takes into account prices) over one using market exchange rates. Eg., according to the dollar/cuban peso market exchange rate, we might conclude that Cubans live on the equivalent of $20 USD per month. Anyone think that tells us very much about the Cuban standard of living?
michael
Yes this is where most people get drawn into the PPP : the per capita GNI (or GDP) numbers look so low. And they are low, if we think of measuring "living standards" which GNI or any of the national accounts do NOT, they only are a ticker to the market economy without double accounting. Comparing national accounts is only a 'market economy to market economy' basis.
So the developing country's GNI per capita feels "low" for various reasons. For example conventional National Accounts would normally leave out all the non-market income of the "traditional" economy in a developing country. Likewise, the whole pattern of life changes to facilitate the lower standard of living. I recall when it was not rare in Europe not have a refrigerator (or only a small one). People ate perfectly well because things were organized around this: small portions at regular prices, distribution channels were structured so people could shop for food every day, etc. Today, if you can't afford a refrigerator it becomes hard to imagine how most West European families would cope.
People think of the low per capita income in developing countries and say it makes no sense in terms of comparing lives. It doesn't, but that is because national accounts compare market economies not "living standards". Along comes PPP which recalculates the numbers and since it moves developing countries closer to the developed countries, people think it makes more sense. But the more "realistic" numbers are entirely coincidental - they still are not measuring living standards because national accounts measure something else. But, on its own terms PPP probably "overshoots" the developed\developing country relationship for narrow statistical reasons (the PPP authors admit this) and distort rankings within developing countries (for example the PPP "conversion" factor for Venezuela is relatively small because it is already much integrated in the tradable economy). Above all the PPP becomes treacherous when one starts to then use them outside of per capita comparisons - e.g. growth rates over time, response to neoliberal measures, etc.
[BTW: I don't know how Cuba's national accounts are calculated. The World Bank does not publish any figures at all. I imagine it is largely guesswork by whomever you are citing (UN?); as you know most planned economies used Net Material Product as their equivalent. There can't be a logical conversion factor for the same reasons PPP doesn't work (apples and oranges). In fact, that is how this international comparison business got started (for example Gerschenkron, Alexander A dollar index of Soviet machinery output, 1951). It was quickly grasped (a bit like PPP) as an ideological tool, ultimately with people like Wolfowitz and Pipes jumping in.]
Paul
