It's not a policy issue, just a statistical artifact. From variance alone you'd expect the US to be middle of the pack, and smaller countries to make up the ex

It's not a policy issue, just a statistical artifact. From variance alone you'd expect the US to be middle of the pack, and smaller countries to make up the extremes on any metric you can make up. Particularly if, as in the case of the OECD, you're essentially cherry-picking high performing, smaller countries. You can't usefully compare ratios where the denominators differ that wildly. If you want to measure the difference between Canada, nordic countries, the US, etc., you need to cluster them into reasonably-chosen and mostly uniformly-sized chunks, collect your metric on each chunk, and compare the distribution of chunks across systems of government/countries/etc. This would actually result in a statistically valid comparison. That's a lot more work and as far as I know nobody ever does it.