Is the Census the New "Voter Fraud"?

A focus on the 2010 census seems to be one of the legacies from Judd Gregg's short time as Commerce nominee. When people went looking for just what the commerce secretary does, anyway, the census was one of the few consequential things they found. And consequential it is.

Having the Obama administration make clear that they wouldn't let Gregg control the census gave Republicans one of the excuses to kick up a fuss for which they are so perpetually looking. And so it was that the census has become "voter registration fraud" 2.0. Or maybe 5.3. Who can even keep track?

Why is the census fertile ground for claims of Democratic skulduggery? Naturally, because it's a way Republicans have long tried to move elections in their favor -- and as the basis for congressional redistricting, the census will have a powerful influence on elections over the next decade.

Amy Sullivan explains the stakes:

As mandated by the Constitution, a census has been taken by the government every 10 years since 1790 in an effort to count every person living in the United States, both citizens and non-citizens. In recent times, the Census Bureau has arrived at a final count by relying on people to mail back surveys and then sending out census takers to go door-to-door in an attempt to fill in the gaps. Those census takers not only visit homes that have not returned the survey, but also seek to count those with no fixed address and those who live in nursing homes, prisons, shelters and other non-standard housing.

The problem is that it is not easy to count every person in the United States, and some communities are disproportionately left out of the total. The 1990 census missed an estimated 8 million people — mostly immigrants and urban minorities — and it managed to double-count 4 million white Americans. Recent or illegal immigrants are often reluctant to answer questions in a government survey, and many experts fear that concerns about government misuse of personal data post-9/11 could hamper participation in the 2010 census as well. Children have also traditionally been underincluded in census totals.

Republicans have long fought to continue these patterns.

Got that? First you undercount populations that tend to vote Democratic. Then you overcount populations that are a bit more likely to vote Republican. Then, when someone proposes a way to arrive at a more accurate count, you accuse them of trying to fix things for the Democrats.

All too often this is done with the aid of reporters like Sullivan -- who, despite itemizing the 1990 over- and undercounts quoted above, nonetheless writes that:

In very general terms, Republicans would prefer to err on the side of undercounting and Democrats would prefer to err on the side of overcounting.

It's a tactic similar to the Republican party's repeated efforts to claim that fraudulent voter registrations will lead to stolen elections:

But if these false registrations don't lead to false votes, and don't affect elections one whit, why do Republicans bring them up again and again?  What makes this political urban legend worth firing US attorneys?  Why would they rather see Alberto Gonzales' neck on the chopping block than give up this favorite belief?

Because they need it.  Without the false flag of voter fraud, Republicans would have no cover for their own extremely direct and well-documented efforts in voter suppression.  They want to keep groups like ACORN from registering thousands of minority voters, so what better way than to show that a handful of workers for these groups have engaged in "fraud?"  They want to justify restrictions that would make it much harder for people to vote.  How can they do this unless they generate evidence that such restrictions are required?

So when you hear about census politics as played out in sampling and statistics, remember that there's something really important there. And as usual, it's about Democrats wanting more voices to be heard, and Republicans wanting fewer.

(Unfortunately, you can't really say national Democrats want all voices to count -- there's hardly a drumbeat around the injustice of prisoners being counted where the prison is located, even when they cannot vote. Political power is thereby shifted out of cities a majority of prisoners are from and into the frequently rural areas where prisons are located. This is not something you're likely to hear a lot about even from advocates of sampling.)