Connecticut is right to count prisoners where they’re housed

Journal Inquirer

By Joe Markley

Prison inmates are among the most permanent of a town’s residents. They reside in a given correctional facility, day and night, often for several or many years.

The U.S. Census uses a common-sense standard to define where you reside. It’s where people “eat and sleep most of the time.”

In the case of an incarcerated prisoner, that’s jail, which is why prisoners are counted as part of the population of the town in which they are located. In reality, many prisoners have only the most tenuous ties to any prior residence.

Logic aside, a new Connecticut lawsuit seeks to use the “last known address” of inmates as their official residence, not the prison.

The stated goal is to put an end to so-called “prison gerrymandering,” but the real motivation is pure politics: The purpose is to increase the stated population (and therefore political clout) of urban areas.

Proponents point to statistics culled from the “most recent addresses” reported by inmates to prison officials upon their incarceration. The data shows about half of all prisoners claimed residence in one of five Connecticut cities: Hartford, New Haven, Bridgeport, Waterbury, or New Britain.

In the public statement announcing the lawsuit, Derrick Johnson, president and CEO of the Connecticut NAACP, raised the ante: “This practice is one aspect of a war of voter suppression targeting communities of color.” That’s quite a charge to make relative to a long-standing, straightforward way of counting people every 10 years.

Here are a few facts about our living patterns which illustrate why counting prisoners as residing at their “last known address” is folly.

More than one in three people live in rental housing in Connecticut, according to Census data. Renters live in their residence for a shorter time than homeowners — the average is just 2.1 years, compared with 8.2 years for people living in owner-occupied housing units. The Current Population Survey shows that 33 percent of people in rental units move every year, and 10 percent of homeowners.

Why is this important? Because it’s readily apparent that even those who don’t reside in jail are transitory. Many of us change our residences frequently, a critical fact that this lawsuit conveniently ignores.

To illustrate, let’s assume John Doe has been renting an apartment in Windsor. He’s found guilty of a crime, sentenced to 10 years in the prison, and sent to Enfield. Doe reports his most recent address as an apartment in Windsor to prison officials, but the Census begins to count him as a resident of Enfield, since that’s where he will “eat and sleep most of the time” — indeed all the time.

Seven years pass. Today, Doe the inmate is still counted as a resident of Enfield. Meanwhile, according to national statistics, the apartment Doe rented in Windsor has probably cycled through three different residents, since the average renter stays in a unit only two years.

Under what stretch of logic are we to assume that Doe will return to that same apartment in Windsor, his “last known address,” upon his release? Don’t wrack your brain. Doe won’t do that.

In truth, it’s only fair that communities that serve an important public need and host a prison within its border derive some level of benefit. Prisons occupy prime real estate in many communities. They are served by local fire and police departments, and municipal utilities. State-run prisons, as public facilities, don’t pay taxes, and they are certainly not known to enhance the quality of life in surrounding areas.

To the extent there are any benefits to a prison in the neighborhood, one is the simple acknowledgement that there is a resident population that has needs — and those needs must be served.

For better or worse, inmates are indeed that town’s residents, albeit temporary, and they are properly counted as such.

State Sen. Joe Markley represents Connecticut’s 16th District. He is serving his fifth term representing Waterbury, Southington, Wolcott, Cheshire, and Prospect.