But your question about number of cell phones is exactly what makes comparison of different economies so difficult. As I see it, there should be a three step process.

1.) Define some kind of metric based on things that you can measure: GDP, life expectancy, number of cell phones, number of vacation days per year, or whatever.

2.) Decide whether that metric is measuring something desireable or undesireable.

3.) Adjust whatever economic controls you have available to attempt to either increase or decrease the metric.

So if you think that the so-called "unemployment rate" metric is an indicator of something that is "bad," then you would work to make it lower. If you think that the per capita GDP is "good," then you would work to make it higher.

To make this work, the very first thing that must be accomplished is the definition of metrics that measure things we are interested in. The current debates are about things like "unemployment rate" (argued against by Krugman in one instance because of discouraged workers), "per capita GDP" (which measures things like cell phones), and "health care ranking" (which takes into account how much is spent--as a negative--but fails to account for massive failures of the system). And then when those metrics are defined, they must be measured over carefully chosen time periods that do not put undue emphasis on things like local recessions.

What are those fair metrics?