Census expert Prof. Persily suggests allowing data users to decide where prisoners should be counted.

by Peter Wagner, October 4, 2004

In previous articles, I have called for the U.S. Census to count prisoners at their home addresses rather as residents of the prison town. All states currently rely on the U.S. Census for data for their legislative redistricting, despite the fact that most states define residence for prisoners to be their pre-incarceration address.

From a voting rights perspective that looks at how Census counts of prisoners dilute urban and minority voting strength, changing where the U.S. Census counts prisoners appears to be the only way to give state legislatures the data they need to draw their districts fairly.

In a paper at a Census Bureau symposium in March, Professor Nathaniel Persily suggested another solution. While including prisoners within the prison town is clearly bad for legislative redistricting, there may be other policy purposes where the number of people physically present is most important.

The solution? Change the form given to prisoners, and ask not only for the prison’s address, but for the individual’s pre-incarceration address. The Census Bureau could then package a dataset for redistricting purposes that would use the pre-incarceration address of prisoners. Professor Persily wrote: “In doing so, the Bureau would remain true to its mission as information provider without having to take a stand on exactly where states and localities should place prisoners.”

This compromise would be a marked improvement over the current Census methodology that denies states seeking to draw fair districts an accurate answer to the question: Where does our population reside?