This gives the impression that the upper-middle class support wealth redistribution at higher rates than the lower-middle class. I'm open minded but I would be quite surprised if that were actually true. That doesn't mean there is nothing to explain-- the lower-middle class still seems to oppose measures that would objectively make them richer at surprisingly high rates. But it isn't as if self-interest doesn't explain a great deal of voting behavior.

Some things to consider:

1. Does opposition to redistribution usually increase during bad economic times? The answer is no for many examples that come to my head. But what I do know is an effect of a bad economy is animosity toward whomever is in power. It is plausible the recession simply correlates with "disputing positions associated with Barack Obama".

2. Ideological stickiness or coalition logic

Now, one of the most interesting things about pre-fabricated political identities is that they come as package deals. There is no logical connection whatsoever between supporting a woman's right to abort an unwanted fetus and supporting subsidies for alternative energy. The strong cultural correlation between these stances creates an illusion of ideological coherence. Since most of us aren't political theorists, we tend not to see that the force determining the various planks in our favoured party's platform is the drive to craft a winning coalition cobbled together from diverse and sometimes conflicting interest groups, not Truth.

...

I think the paradox, or the irony, is that the evolution of partisan coalitions can lead to bizarrely incoherent partisan worldviews. Easy money in a recession is the objectively pro-business position. However, the rising preeminence on the right of the idea that inflation, like taxation, is largely a mechanism of unjust big-government expropriation can, through mere association, make this viewpoint seem like the "pro-business" one, even if it isn't. It's this kind of drift in the composition and ideology of partisan coalitions that can make even debate over economic policy seem like just one more front in the culture war.

1. The Farmer-forager binary. In particular consider that fear made farmers. If egalitarianism is associated with forager thinking then fear caused by economic uncertainty would trigger farmer-type anti-egalitarianism and an emphasis on self-reliance. Those with the most precarious economic situations are the ones most likely to oppose egalitarianism as the economy gets worse.

   In particular, the farmer-forager thesis explains the following puzzle that the signaling theory does not. If opposition to welfare is driven by people signaling that they don't need it we should expect the people most likely to oppose welfare to be those most likely to be mistaken for being on welfare. In particular, we should expect lower-middle class African Americans to be especially worried about looking like they're on the dole and therefore especially vociferous in their opposition to welfare. But we in fact we find the opposite: African Americans overwhelming support welfare programs relative to whites. And this response is exactly what one would expect with the "more for mine" attitude associated with farmer norms. Fearful occasions of resource uncertainty cause us to shrink or sphere of moral concern, be less willing to share and be more suspicious of strangers especially those who look different. Also note that the rates of redistribution in more homogeneous European states-- particularly Scandinavia -- vastly outstrip rates in more diverse countries like the US.