Demand curves are downward sloping even for prison guards

This appears to be a pretty big difference in rates between male and female prison staff having sex with prisoners -- more than a two-to-one difference.

Inmate Michael Murphy usually started by seeking a small favor. That would often lead to a kiss or love letters. And in at least five cases, he convinced female prison employees to have sex with him or do other illegal favors.

In each of those cases, the female corrections employees were caught, shamed and forced out of a job, according to documents detailing an investigation by Montana prison officials and obtained by The Associated Press after an open-records lawsuit.

The female officers described Murphy as the aggressor, even as the predator. But that makes no difference in either state or federal penitentiaries, where prison employees — male or female — are the violators if they have sex with inmates.

A Justice Department study shows that cases like Murphy's are common: Female staff are more often implicated than their male counterparts in prison sexual misconduct. While many cases could be considered consensual, incarceration experts and female prison guards say the problem is much more complicated.

In some cases, the women reported that they couldn't say no to the inmate out of fear, or were afraid to go to a co-worker out of shame at what had happened. One small mistake often led to something else. . . .

Martin Horn, now a professor at the John Jay College of Criminal Justice, said female workers who have sex with inmates are often treated less harshly by officials than male worker who do the same.

"As long as we have a double standard we are going to see these kind of behaviors," Horn said. "It is a very slippery slope we go down if we say we are not going to hold female officers to the same standard."

A 2007 U.S. Department of Justice study analyzing the prevalence of sexual assault in state and federal prisons found that 58 percent of staff perpetrators of sexual misconduct were female. . . .