I'm not sweating the tightening generic ballot. It probably reflects one of two things. First, some polls have a too large number of self-identified Democrats in their voter sample. The second is some Democratic leaners coming home. The second factor is more relevant but should be kept in perspective. The Democrats are going to get their 45% or more of the House vote. I'm confident that Republicans will win control of the House and win 9 to 11 seats in the Senate. I think Angle will win but the Democrats will win close ones in Connecticut and California.
I'm watching the California Senate race because it could be a sign of where politics is going. Republicans should be winning this race. The Republicans have the lousy economy working in their favor. Boxer is an extremist (she is cagey about whether it should be legal to "abort" fully delivered fetuses...ugh...babies), and she has an obnoxious personality. Fiorina isn't a perfect candidate, but she is competent. Fiorina's problem seems to be that Republican gains seem concentrated among two groups. First are right-leaning voters who (if polls are to be believed) are very excited about voting. The second group is made up of white persuadables who don't have a strong liberal/Democrat political identity. The Republicans are getting huge margins among whites. Among white voters this isn't 1994. This is 1984 at the congressional level. In other words, the white voters who aren't voting Republican are committed liberals. That is enough for Republicans to win in most places in 2010, but it might not be enough in California. The Democrats are, under very adverse circumstances, doing surprisingly well among Latino and African American voters. Obama's job national job approval rating among Latinos is 55%. I'm not sure exactly how much stock to place in the California polls, but Boxer seems to be beating Fiorina among Latino voters by about 2 to 1. The problem for Republicans nationally is that the national electorate will look a little more like California's every election cycle starting in 2012. Or to put it another way- if ethnic voting patterns stayed the same, Scott Brown would probably have lost his Senate race in the demographic Massachusetts of 2020.
Now there is no guarantee that the racial or ethnic voting margins of the present will reproduce themselves exactly in the coming decade-plus. But there are reasons to worry. The Democrats are holding undivided power and the American labor market is as bad as anyone under seventy can remember. Unless something changes, the Republicans are near their historical ceiling among white voters and are only barely back where Bush was among Latinos in 2004. The situation points to Republican losses among both groups if circumstances improve even a little.
So that means that Republicans are going to have to earn (or not) gains among Latinos, and probably under less favorable circumstances than they have now. It isn't exactly clear how to go about doing that, but Ricochet's Rob Long asked an interesting question. I'll try to give a tentative answer tomorrow.