Jenny B:
> Except it leaves the mystery of (1) why anyone in the lower
> 90% of income votes for Bush and (2) why people vote at all.
> I mean, what motivates them, not 'what's the point?' Because
> a large percent--often a majority--don't vote.
> So voting as an act needs to have some more motivation than
> the empty exhortations 'don't forget to vote!' we get around
> election time. I still think Frank is essentially right when
> he says that the Democrats could do a lot better among the
> U.S. working class if they weren't Republicans light on
> economic issues. For example, they could actually make a
> case for why the estate tax is a good thing, and maybe should
> be higher, say, confiscatory. Let's have a public debate
> about whether the children of the rich should be
> automatically rich or whether they should have to get a job.
> But that's against most politicians'
> personal interests, so we'll never find out if it's against
> their political interests.
Good question. My answer to it is to adopt the correct behavioral model. There is a general tendency among social scientists (especially polit-sci and economists) and also left-wing social commentators to adopt the rational choice model which stipulates behavior producing maximum utility. Therefore, people are baffled when facing persistent behavior that does not seem to produce much utility. Voting seems to be one of them.
However, being a sociologist gives you a smorgasbord of behaviors models that do not make the rational choice or even the mentalist (i.e. states of mind leading to behavior) assumption. With that in mind, I suggest two behavioral models of voting.
First, voting is a form of symbolic interaction - a form of participatory ritual that connects people to each other and creates a sense of a social bond and belonging - like religious rituals (dances, feasts, prayers) cultural rituals (going to cultural events, concerts, performances, sporting events, etc.), family rituals (Turkey day dinners, family reunions) etc. In the same drift, people vote because they feel the need to be connected to the larger social entity called America (or Germany, or Mother Russia, or any other nation) - even though that sense of belonging is the only gratification they obtain. This explains why in countries where the utilitarian outcomes of voting is close to nil (e.g. in one party states) voter participation is still high, even higher than in the US.
Conversely, people who have no need to be connected to a larger community or nation do not bother to vote. They may feel the need to be connected only to a much smaller local community or even family and thus engage only in those social bonding rituals that are appropriate for those local or familiar communities (e.g. anything from outings at the local fire company, to bowling, hunting, family picnics, etc.), or perhaps have no need to be connected at all, in which case they do not engage in any rituals either national (voting) or local (bowling, picnics). I believe that due to a highly developed individualism in the US, the latter group is much larger than in most other countries, hence the voting participation is much lower despite its relatively high utility (e.g. comparing to one party states).
Another behavioral model that I find useful comes from the "garbage can" theory of organizational behavior. This model basically holds that in situations with multiple and frequently changing stakeholders, the most likely outcome is that that the least objectionable to the particular set of stakeholder that happen to be involved at any particular time. However, and equally importantly, the parties would ex-postfacto justify that outcome in terms of their prevailing ideologies. For example, the company management make decide on a course of action that is least likely to be opposed by the various stakeholders, but will justify that decision in terms of maximizing efficiency and competitiveness that form a core belief of managerial ideology. Likewise, voters may cast their vote on the person whom they perceive, for whatever reasons, as the most likely winner, which provides them with self-serving ideological assurance that they selected the best candidate for the country.
In both cases, people participate in elections to obtain different emotional gratifications from the participation in the process itself, rather than from the outcome of that process. Therefore, it does not matter than much what actual policies the candidates propose - as most people have no time, skills, or both to evaluate the costs and benefits of those policies. The only things that matter are whether a person has a need to belong to a larger social entity (a state or a nation) and the likeability and sex appeal of the candidates. Stated differently, voting is no different than participation in any other popularity contest or pageant. The voters have no stakes in the outcome - it does not really matter to them whether Miss X or Miss Y wins. What matters is whether participating in the pageant "connects" them to others, and whether the winners are likeable.
Wojtek