However, I think the 2008 data makes the point even more strongly that using the ratio of public to private employees as metric for “big” government is meaningless. The implication of the table is that somehow political decisions and philosophy about government influence the ratio that the RGF has chosen to use as a valid metric (public to private ratio).

I just broke out the 2008 list of states by political party (from 2008 election–one of the few times when there were roughly the same number of states voting for each party, and where there was a clear ideological choice. I did so because conservatives can sometimes be Dems, and moderates (though rarely liberals) are sometimes Republicans).

Of the 15 “worst” states with the highest public:private employee ratio, 2 voted Democratic (NM and WA) and 13 GOP. Of the 15 “best” states with the lowest public:private employee ratio, 3 voted Republican (FL, TN, and MO) and 12 voted Dem–including such liberal basions in the top six like Massachusettes, Pennsylvania, Rhode Island, and Illinois.

Further, the 15 “worst” are overwhelmingly rural, and contain an oversize share of conservative inter-mountain, plains, and southern states. In order, they are: Alaska, New Mexico, Wyoming, Mississippi, Oklahoma, West Virginia, Kansas, North Dakota, Louisiana, Alabama, South Carolina, Montana, Idaho, Washington, and Arkansas.

So is the RGF’s thesis now that Republicans, southerners, rural people, and mountain west folks are for big government, while New Englanders, Democrats, and city folks favor small government? Is the Rio Grande Foundation really saying that Massachusetts, Rhode Island, and Illinois are great small government states, and that Wyoming, Kansas, and North Dakota are bloated big government states?

The statistic was meaningless in 1970, and is just as meaningless, maybe more so, using 2008 data (remember, over the last 40 years there has been a clearer alignment between geography, ideology, and party, so the fact that the current data shows such a strong correlation between southern, mountain west, rural, and Republican states and “big” government is even more meaningful).

But I do honestly appreciate Paul’s pointing out that I was using the 1970 data (and apologize for my mistake), provding more links to the correct data, and engaging in an interesting debate over how to gauge the size (and appropriateness) of government.