In my view the U.S. ought to encourage the development of entrepreneurial states rather than other forms in order to avoid international conflict, including sabotage, industrial espionage, and even armed warfare. I speculate that leadership for this move is likelier to come from the leaders of multinational corporations and nongovernmental organizations (NGOs) than from leaders of the national security apparatus and the political establishment, but I concede that business leaders are generally not prepared for such a role today.

One thing I didn’t mention is how a voter could actually feel the pull of two or more categories. Say you have an Issues-based voter whose surrounding family and friends are Party-based. This could influence his decisions. He could feel conflicted at the ballot-box, which could compromise his rationality.

Another thing to consider is how politicians campaign to each group of these voters simultaneously. Politicians try to look morally upstanding, identify with voters, discuss big issues and fight for the common man all at the same time. A politician may also frame his campaign to appeal to one category if he feels that the majority of voters lies there. For example campaigning on the Iraq war with the idea that more people are against it than for something else, like stem-cell research.

Although I can’t remember which document, I remember reading about the US founding fathers who thought of the voting public as people who vote based solely on their own interests. They based their democracy on the idea that if everyone voted for their own interest, all interests would be represented proportionally on a national scale which would be “fair”. It seems to me they didn’t see that there are many different types of voters which can unbalance their imagined system.

Adrian: I could easily see identity-voting as a variant of one of the others depending on the specific nature of the thoughts involved. To whatever extent black interests differ from those of other races, a black person focused on those distinctive issues would be an interest voter. A person (black or otherwise) who sees the problems of the black community as problems for the nation as a whole (not necessarily a bad idea)would be an issues voter. A person who sees one party as being far-and-away the best for black people would be a party-voter. A person who votes for black candidates whenever such are available would be a personality-voter.
Chief: I wonder how many people, if presented with information on three candidates stripped of all known bias inducers (pictures, names) or ‘irrational’ information would even be able to make a decision between the three?

Like neoclassical economics, much democratic theory rests on the assumption that people are rational. Here, too, empirical evidence suggests otherwise. Voters, it has been demonstrated, are influenced by factors ranging from how names are placed on a ballot to the jut of a politician’s jaw. A 2004 study of New York City primary-election results put the advantage of being listed first on the ballot for a local office at more than three per cent—enough of a boost to turn many races. (For statewide office, the advantage was around two per cent.) A 2005 study, conducted by psychologists at Princeton, showed that it was possible to predict the results of congressional contests by using photographs. Researchers presented subjects with fleeting images of candidates’ faces. Those candidates who, in the subjects’ opinion, looked more “competent” won about seventy per cent of the time.

When it comes to public-policy decisions, people exhibit curious—but, once again, predictable—biases. They value a service (say, upgrading fire equipment) more when it is described in isolation than when it is presented as part of a larger good (say, improving disaster preparedness). They are keen on tax “bonuses” but dislike tax “penalties,” even though the two are functionally equivalent. They are more inclined to favor a public policy when it is labelled the status quo. In assessing a policy’s benefits, they tend to ignore whole orders of magnitude. In an experiment demonstrating this last effect, sometimes called “scope insensitivity,” subjects were told that migrating birds were drowning in ponds of oil. They were then asked how much they would pay to prevent the deaths by erecting nets. To save two thousand birds, the subjects were willing to pay, on average, eighty dollars. To save twenty thousand birds, they were willing to pay only seventy-eight dollars, and to save two hundred thousand birds they were willing to pay eighty-eight dollars. What is to be done with information like this?

The article is in The New Yorker.