Search This Blog

17.5.07

Losing whites bigger worry for LA Dems, data show

With the release of the latest census estimates, debate re-emerges over the impact of the hurricane disasters of 2005 on 2007 elections and beyond. As I have demonstrated elsewhere, Republicans can expect to get about a 50,000 vote boost relative to Democrats for statewide offices this fall. Do census data confirm this trend, and what will cause it? If you believe the conventional wisdom, you may be in for a surprise.

One notable thing to ponder is that the July 1, 2006 estimated white population (which includes those of a mixed-race background) actually shows a small increase over the same estimated July 1, 2005 population of about 4,000, whereas black population has dropped significantly by almost 139,000, over 9 percent. As far as elections go, this doesn’t quite tell enough, because we need to estimate how this affects the proportion of registered voters and of those who actually vote. Further, we need to compare by race, since the majority of whites vote Republican, and the vast majority of blacks vote Democrat.

If we look at registration totals, there hasn’t been that big of a loss in registrations from 2004 to the 2006 special elections despite disasters. The dropoff of both whites and blacks was around 43,000 compared to the 2,000 or so gain between 2000 and 2002, but of that dropoff, about 31,000 were whites. This is consistent with the trend of recent years: white registrations are losing ground relative to blacks’. Thus, all things equal, blacks voting power would be increasing.

But because of the storms, things are not equal. Further, there is the persistent turnout gap between races that one might think would be aggravated by this recent history. Statewide, over the 2000 federal, 2002 federal, 2003 state, and 2004 federal elections, the gap between the races in turnout averaged 8.75 percent. Yet in 2006 it was below this average, 7.9 percent. Remarkably, this was despite a huge dropoff caused by displacement in Orleans Parish among blacks: there, whites turned out 4.6 percent less than the statewide white average of 27.6 percent, but black Orleanians were 12.8 percent below the statewide black average, at a miniscule 6.9 percent. (In the past, black Orleanians typically voted at slightly higher rates than blacks statewide.)

This obviously means that blacks in other parishes were taking up the slack, so to speak, for those in Orleans in rates and thus to some extent numbers – but not totally. Consider also the proportion of the statewide electorate whites comprised – in 2002, blacks made up 26 percent of those voting falling 2.1 percent from that in 2006. This is consistent with the estimated population change that also saw a 2 percent larger gap open up white over black.

So, we have an overall white-black turnout rate gap holding steady, but relatively more whites and fewer blacks – even as the proportion of black registrations increases statewide. The latter is reflected in the estimated registration rate for whites in the state of 67.2 percent and for blacks of 62.3 percent, compared to 69.5 percent and 57.5 percent respectively the previous year – a difference more than halved.

To put all of this in perspective, this means that Democrats perhaps should be less concerned about retreat of its black base and more about white flight from the state and from their party. Even though the black base is down, the numbers indicate disproportionately it is non-registrants that have left the state, blunting this reduction. The opposite seems to have happened for whites: those eligible to be registered have disproportionately taken off. Further, turnout rates, even taking displacement into account, don’t seem to have suffered relative to whites.

Note that in my paper referenced above I surmised that the state would have lost about 135,000 voters for 2007 while the census figured a population decline of about 194,000 – in other words, using the registration rates above, very well predicted as of today. But of those 75,000 or so Democrats projected lost, adjusted for the figures above only about 40,000 of them are black. Republicans project only about 26,000 whites lost – 9,000 below the Democrat total.

In short, while Democrat losses among blacks are worrisome enough for them, statewide white departures will hurt almost as much in 2007. Further, since a disproportionate share of whites who left and won’t return were voters while blacks were the opposite, the nature of that kind of exodus going forward really hurts the party.

No comments: