Who Gets Cheated by an Abrupt End to the Census

Published by Michael Frank on

It is—and isn’t—who you’d think

Article 1, Section 2 of the Constitution requires a decennial census. It is the basis for  reapportionment, the drawing of legislative district lines for all levels of government from the federal House of Representatives all the way down to local city councils and some towns and villages. The census has also come to be used for an array of other purposes, arguably most significantly the distribution of funds to states and localities for large numbers of federal programs. In other words, what’s at stake in the census is who governs and who gets what—and how much—from the national government. Notably, this past July, despite the absence of Constitutional language to defend it, the Trump Administration issued a directive to exclude undocumented immigrants. And then in August, the president proposed ending the count a month early, in September rather than October.

That attempt has been struck down in court, as has discounting non-citizens, at least for now. 

So if you’re reading this post and still haven’t completed a Census form, there’s still time. Please #FillItOut: Just click here.

As the New York Times recently reported, ending the Census early, or any sort of under-count, can strongly impact communities of color and the poor. The Times’ reporting also makes clear that, especially as the Census Bureau this year pushed for greater online counting (meaning, filling out a form on the Internet), the damage of ending the count early will also disproportionately be felt by anyone less likely to have online access. Lack of broadband access in rural areas of New York State was a serious concern even before remote back to school this year again drew attention to this digital divide.

And it is not just a rural problem; disconnected households tend to be poorer, and/or, elderly. 

Consequently, killing the count early may not result in the scale tilting President Trump seems to perceive; the Times’ reporting leans heavily on research by the CUNY Mapping Service at the Center for Urban Research at the CUNY Graduate Center. That data, along with research by the American Statistical Association (see chart), as well as current Census Bureau reporting clearly shows that states more heavily impacted by an early cutoff or undercount largely leaned or voted Republican in 2016. These include Alabama, Mississippi, Montana and South Carolina. Also, states like Florida, Louisiana and Georgia could potentially each lose more than $50 million a year in federal Medicaid funding. Florida could lose a congressional seat, as could Montana.

The impact of an early halt to the enumeration will also be felt closer to home. New York is a donor state, meaning that we send more in taxes than we get back from the federal government. Yet even with New York reaching 96.7 percent of its targeted tally (percentage is of housing units) as of September 22, according to the Bureau, portions of Ulster, Dutchess, Greene, Sullivan and Orange counties in our region all contain swaths of what the CUNY mapping project identifies as the hardest to count census tracts in the nation. Approximately 36% of New York’s current population (7,091,051) lives in hard-to-count neighborhoods. They had anywhere from a zero to less than a 73% Census return rate in 2010.

CUNY’s data shows, for instance, that Sullivan County’s self-response rate (by mail or web) as of early August was 9.5 points behind its final 2010 rate, and Greene County was 9 points behind. Some counties in our region do look better on paper. For instance, Ulster is over 4 points better than in 2010. But the regional map, where portions shaded in orange and red indicate hard to count sections, makes it clear that regardless of overall progress, more rural parts of the Hudson Valley are more endangered by an undercount.

There’s also a misunderstanding of who actually gets impacted by an undercount. Not counting undocumented immigrants doesn’t just harm them. It reduces the funding available to schools of, frequently, their American-born children—and of every child in those schools, because an entire district gets denied monies, not specific households. In other words, shortening the time we have to count people or not counting certain people because folks in Washington wish they weren’t here hits unintended “collateral” targets. In New York that also means more rural parts of the state, again because of lack of Internet access and poverty, are financial losers from lower totals. The state map shows areas of New York that trail the 2010 Census (orange and red are very far behind), and these are largely rural districts.

This isn’t strictly causal; in a year with a pandemic overlapping fairly tremendous disparities in wealth and digital bandwidth, it’s exceedingly difficult to know at this very moment what effect each aspect has on Census follow-through.

Discounting knowing the exact cause, the net impacts are real. 

As noted, in addition to determining the size of the states’ House of Representatives delegations, the Census data is used to redraw all legislative districts at all levels of government to ensure that they have roughly the same population. This process is called redistricting, and it is the foundation of our representative democracy. Who is counted, and where, will help to determine what the new Congressional District lines look like. Undercounts will severely disadvantage not only the state because of decreased representation in Congress, but also the areas within the state that are undercounted.

Historically undercounted minority populations will be impacted in redistricting at the local level as well. An undercount in the Hispanic population in a small city, for instance, might allow a redistricting commission to split the Hispanic population among several districts, thereby diminishing their political influence. A racial-ethnic population that is fully counted can make up the majority in just one ward, and the voting rights act would protect their representation, but if there’s an undercount, the threshold might not be met and federally required Voting Rights Act parameters could be ignored.

An accurate Census count is at the heart of our government. It ensures fair and equal representation, and will set the stage for elections for a decade (or more). The impacts will be far reaching, and across all demographics, regardless of political persuasion. In other words, everyone should fear the unintended consequences of monkeying with what our founders rightly viewed as not just an apolitical process, but a patriotic one. At least as of this writing, you have 33 days left. Fill out the Census.


Leave a Reply

Avatar placeholder

Your email address will not be published.

%d bloggers like this: