Prisoners placed 100 miles farther from home get 60% fewer visits and reoffend more

In Washington State, a 100-mile increase in home-to-prison distance cut monthly visits by 60% and raised one-year readmission by 9%. One extra visit a month lowered three-year re-incarceration by about 7% and post-release housing voucher use by 12%.

Illustration of a long road stretching from a family home to a distant prison on the horizon.
Share this:

Sending someone to a prison 100 miles farther from home cuts their family visits by more than half and makes them more likely to end up back behind bars. Using records on more than 20,000 people and 200,000 visits in Washington State, an economist estimates that the extra distance raises the chance of returning to prison within a year by 9 percent.

The study is the job market paper of Danielle Nemschoff, a University of Chicago economist. Its latest version, dated July 31, 2026, is available on her website.

An assignment system blind to family ties

Washington’s Department of Corrections places people in facilities based on security classification and available beds. It does not consider where their families live. That rule creates something close to a natural experiment, since two otherwise similar people can end up at very different distances from home.

Nemschoff linked home addresses, facility assignments and full visitation logs. She found no sign that distance was tied to access to programs, education or prison jobs, or to traits such as age and offense once security level was accounted for. In Washington, a 100-mile trip typically means 1.5 to 3 hours of driving, and public transport to prisons is limited.

Fewer visits, more returns

Each extra 100 miles lowered the probability of receiving any visit by 11 percentage points, from a 44 percent baseline. Monthly visits fell by 0.6, a 60 percent drop from the average of about one a month.

Readmission to prison rose by 0.8 percentage points within one year of release and by 1.1 points within three years, increases of 9 and 4 percent. Using distance as a lever to isolate the effect of visits, Nemschoff estimates that one additional visit per month cuts one-year recidivism by about 14 percent and three-year recidivism by about 7 percent. She notes that one more visit a month doubles the average, and scales it down: a visit every two months lowers three-year re-incarceration by roughly 3.5 percent.

Housing and time served

Visits also seem to help people leave on time. An additional monthly visit reduced the fraction of the sentence served by about 1 percent, consistent with fewer disciplinary infractions, since early release credit can be revoked for misconduct. Counting future stays avoided, that adds up to savings of over $4,000 per person at an estimated $174 per day of incarceration.

The paper points to housing as one channel. Washington issues vouchers to people who lack an approved release address, and an extra visit per month reduced voucher use by 12 percent. Visits in the final months before release, when address checks begin, mattered most.

Simulations of placement rules that favour proximity, within security and capacity limits, predict recidivism falling by 2 to 4 percent. Nemschoff notes that two earlier studies in Iowa and Denmark found no significant effect, and that hers draws on roughly twice the Iowa sample. She also flags that some county jails are replacing in-person visits with video calls, a substitute whose effect on reentry has barely been tested.

Study Details: