1) I was responding in large part to cross-country plots and regressions, e.g., https://randomcriticalanalysis.files.wordpress.com/2014/11/above-expected-500x4

1) I was responding in large part to cross-country plots and regressions, e.g., https://randomcriticalanalysis.files.wordpress.com/2014/11/above-expected-500x406-1.jpg https://randomcriticalanalysis.files.wordpress.com/2014/11/a... 2) It's not sensible to run regressions when observations consist of US and EU-(14|28) alone. Small n and these all of these common EU aggregations would be MUCH poorer than the US. 3) US states have a fiscal, political, legal, social, and other union much beyond that of EU member states in the EU. I would very much anticipate larger spillover effects between US states than between EU member states for this reason (as in, MS's standard of care will be much more like MA's than Slovakia will be like Norway), not to mention gains from greater economic union, transfers, and the like. 3) Perfectly comparable data do not exist to make direct comparisons across different levels of analysis feasible. 4) I actually did run some regressions on US states and found: (i) states vary quite substantially in their expenditure levels and (ii) consumption (PCE) to be an exceptionally stronger predictor than GDP, personal income, and some other economic indicators. DC spends more than twice as much as Utah (and if you don't like DC as an observations, similar results are found for MA) https://randomcriticalanalysis.files.wordpress.com/2016/09/rcafdm_39_hce_by_pce_by_state.png https://randomcriticalanalysis.files.wordpress.com/2016/09/r... Even so, even if I were to try to equate these measures, I would not except the slope within countries like this to match the cross country slope due to some of the aforementioned spillovers. 5) I actually did analyze outcomes between states and even counties. https://randomcriticalanalysis.wordpress.com/2016/11/06/us-life-expectancy-is-below-naive-expectations-mostly-because-it-economically-outperforms/ https://randomcriticalanalysis.wordpress.com/2016/11/06/us-l... 6) Tho the US has greater geographic heterogeneity in outcomes than most of individual EU countries due to its size, greater historic diversity, and more, it's worth pointing out that health systems do not produce equal outcomes within countries. As I pointed out in the above link, the northern areas of England and southern parts of Scotland perform quite terribly. https://randomcriticalanalysis.files.wordpress.com/2016/11/screenshot_1170.png https://randomcriticalanalysis.files.wordpress.com/2016/11/s... These regions, btw, are where a large fraction of people in the US came from (especially those that settled in and around Appalachia). There is a whole literature indicating that "deep history" predicts economic and social outcomes many centuries later.....