Cool.
"The safest countries included Iceland, Japan, Israel, Singapore, and Australia"
Israel? Wow!
There is of course a problem with placing too much emphasis on bivariate correlations: if 55% of prosperity are explained by freedom and a further 85% by good governance aren't you explaining 138% of the observed variation in GDP per capita?
The only problem with it (and I would love to see a government do it) is that you cannot EVER get rid of poverty. Poverty would simply move (numerically) and you'd have new people in poverty to try to get rid of.
Communism (if it actually worked with people) is basically the only way to get rid (completely) of poverty by simply bringing everyone under the poverty line and canceling it 
Rantingkraut, there are of course overlaps between "freedom" "safety and "good governance", they are all the same thing really.
Even if they weren't there's no reason why the correlations shouldn't add up to more than 100%. They are all pre-conditions.
e.g. of successful businesses, you might observe that 75% were set up by somebody with burning ambition; 60% of founders had a bright idea how to do things better; 50% of businesses are where they are because of blind luck and so on (figures plucked out of thin air).
e.g. you can observe that 100% of people have a biological father and also that 100% have a biological mother (as a pre-condition of being born). There is no point adding 100% to 100%.
"The only problem with it (and I would love to see a government do it) is that you cannot EVER get rid of poverty"
This is if you subscribe to the modern version of the word poverty which is really relative poverty. I would suggest that we have wiped out absolute poverty in the UK.
@ Mark,
I don’t think that safety freedom and good governance are the same thing. You can be free without being safe. Two countries can give their inhabitants similar levels of freedom and safety while having different degrees of corruption etc. My main issue here is not with the definition of the variables –although there will be issues- my main problem is with the interpretation of the bi-variate correlations.
Bi-variate correlations like this are almost certain to involve some degree of omitted variable bias: If richer countries are freer as well as safer and better governed then a correlation between GDP per capita and good governance will attribute part of the impact of safety and freedom to good governance –thus overstating its impact (and likewise for the other correlations).
The way round this would be to consider all relevant variables simultaneously –e.g. the three above plus the inputs of production factors like capital, skill, labour etc.
Mark - poverty must be measured as a combination of relative and absolute. both sides have valid points.
Pommey,
Good post will link often and often again please keep it in good order for your future fans,
Thanks,
Tony
"Dahl then looked at lost years of life per 100,000 population due to hunger, disease and conflict in each country. He got an amazing R2 of .51, which he said means that about 51% of prosperity can be attributed to people living in safety"
Or are the people living in safety that can be attributed to prosperity? (specially living safe from hunger and disease?)
thanks Tony.
I'm not sure that looking at the performance of immigrants is the most sensible way forward: Russians and Indians who go and live in America will tend to be those Russians and Indians who were richest, best-educated and most motivated to start with (since your average Moscow ex-army chap or Uttar Pradesh peasant won't be allowed into the US embassy to apply for a tourist visa, never mind given permanent residence, even if they thought it was a possible option and could be bothered to try in the first place).
But that's a minor nitpick. Interesting post...
These regressions are highly questionable - you may be regressing income upon a consequence of income. Many of these corruption indicators and henceforth actually look like outcome variables - a good paper to read is http://
www.economics.harvard.edu...ions_Growth.pdf
it's not definitive by any stretch, but it does illustrate some of the problems using these measures of economic freedom etc.
Pommpgrante, I think these grahps are a load of bulls... The first graph only shows that their is a rough (very rough)between "freedom" (which is a vauge concept) and prosperity, and it also must be said that GDP isn't the best way to measure prosperity. The secong graph is even more rubbish, as it just shows that some countries can have no safety and still be relativly rich. The last graph is interestiung, but it still doesn't justify you theory.
If you would like to see more empirical evidence presented superbly, I've posted a brief piece with links to the Gapminder software here. Professor Rosling's video (at that link) is both entertaining and enlightening.
thanks Steve. Ill take a look.
Commenting by HaloScan
