This is a fascinating and revealing chart illustrating our curious public mood swings. It was produced by Larry Bartels (see here) using data compiled by James Stimson (see here). It measures the “policy mood” of the country since 1950. The higher the score, the more conservative was public opinion at that particular point in time.

Stinson derived this policy mood index from responses to a wide range of public policy surveys. Since it does not rely upon self-identification as liberal or conservative, it is arguably a more nuanced and accurate measure of where public attitudes fall on the liberal-conservative spectrum.

The most obvious thing to note here are the dramatic swings. Clearly public attitudes about major issues are not as fixed as we might imagine. Over the past 65 years public opinion has swung up and down by almost 20 points. Clearly public sentiment can be swayed significantly.

The most interesting thing about these swings is revealed when you refer to the Administration timeline that Bartels added to the x-axis. If you study this a bit you’ll undoubtedly start to scratch your head in confusion. Under each Democrat administration the country became more conservative, and under each Conservative administration the country became more liberal. This is entirely counter-intuitive and immensely important.

The nation became dramatically more liberal during the Eisenhower years. Ike was moderately conservative overall but staunchly conservative on economic issues. It became slightly more conservative during the moderately liberal Kennedy/Johnson era but shifted far toward the conservative extreme during the very liberal Carter years. Similarly, during the extremely conservative Reagan era, public sentiment shot back down again toward the liberal end of the spectrum. Under Clinton, the public then became more conservative and after a year or two of George Bush became more liberal again. Finally, under the very liberal President Obama, we have become dramatically more conservative.

The next observation may or may not be significant, but the swing has been between 30 and 50 on this scale of conservatism. I can’t imagine what kind of views it would require to earn a 100% rating on this scale, but for what it’s worth public opinion has remained solidly on the liberal side of the spectrum. This seems to defy the popular meme that America is a “near Right” country. It suggests we are actually (still) a “near Left” country.

But that may not last if trends continue. Although this chart bounces up and down, there is still a clear upward best fit line. This supports the long-term trend toward conversativism reported independently by other sources. Essentially each President since Eisenhower has rated as more conservative. More specifically, each Republican President has been more conservative than the previous Republican President and each Democrat President has been more conservative than the previous Democrat President. This reinforces the observation that despite these swings, the Conservatives succeed year after year in moving the “center” ever farther toward the Right.

So what conclusions can we draw from these data? First, it isn’t true that we are intractably divided and cannot change. Clearly a very significant fraction of us can be moved a substantial amount in one direction or the other. Next, we are perhaps overall more liberal than the Right would like to have us believe. However, we are trending ever more conservative and that is never likely to reverse as long as liberals keep voting for the “lesser of two evils” who is still ever more conservative than his or her predecessor.

Finally, as Bartels pointed out in his article, Presidents do not actually succeed in shifting public opinion their direction. The data rather show that the public tends to recoil reliably away from the President in their attitudes. Paradoxically and counter-intuitively, these data suggest that the fastest way to shift public sentiment toward the liberal end would be to elect a highly conservative President. And the best way to reverse the long-term trend toward conservatism may be to allow our “lesser of two evils” Democratic candidates to lose.