Another Example of Good Intentions Gone Wrong

Jennifer Doleac

Jennifer Doleac

by James A. Bacon

Last year Governor Terry McAuliffe signed an executive order to “ban the box” prohibiting employers from asking job seekers about their criminal history at the initial job stage. The goal was to “remove unnecessary obstacles” to felons seeking employment after incarceration. How could one object? Once felons have paid their debt to society, we should ease their transition back into the workforce, right?

It turns out that things don’t always work the way we expect them to. From the Daily Progress:

Research published recently by Jennifer Doleac, an assistant professor of public policy and economics at the University of Virginia, found that ban the box policies actually lowered the probability of employment by 5.1 percent for young, low-skilled black men and 2.9 percent for young, low-skilled Hispanic men.

According to Doleac, who conducted the study with the University of Oregon’s Benjamin Hansen, the lowered chance for employment comes from the unwillingness by employers to take chances on hiring someone without knowledge of their potential criminal history.

“Simply taking away information about whether someone has a record doesn’t stop employers from caring about someone’s criminal background,” Doleac said. “It just leaves them to guess based on the remaining information they do have.”

All too often, that “remaining information” is age, race, ethnicity and socioeconomic background. (Hat tip: John Butcher)

Bacon’s bottom line: Society is extraordinarily complex. Political ideologies (both on the left and the right) provide simplified models for how society works. Often those simplified models overlook important linkages and feedback loops that lead to very different results than anticipated. Individuals and private entities can quickly alter their behavior to adjust to reality; government adjusts much more slowly, if at all.

Will McAuliffe rescind his “ban the box” order? I’m not betting on it. The social engineer’s response to problems created by a law or regulation is to “fix” the emergent problem by enacting more laws and regulations… thus creating new problems. 

It’s fine to try new ideas, but we have to pay attention to whether they work or not. If they don’t, we need to reconsider them. Good intentions are not enough.

There are currently no comments highlighted.

23 responses to “Another Example of Good Intentions Gone Wrong

  1. it’s probably a relevant point but employers CAN order a background check – which now days can be pretty quick and pretty cheap…

    and for some jobs… like dealing with kids or transit driving, etc.. mandatory no matter what boxes are checked.

    • Come on Larry, you know the only job of government is to make sure that businesses never spend more time or money than those businesses deem absolutely necessary. Anything else is onerous regulation.

  2. I, for one, would have anticipated that business owners and managers would have responded in the most racist fashion possible, but I can understand why the governor would have been reticent to admit this likelihood out loud.

    But you’re right, it’s better to blame the governor for trying to achieve a pro-social goal with regards to our treatment of people who have served their time than the actual people behaving in anti-social ways.

    • I’m not blaming the governor for the outcome. I’m saying he needs to reconsider his approach. Perhaps there are other ways of accomplishing the same goal.

      • naw… you’re criticizing the approach and leaving it at that.

        I’d be SHOCKED if I starting seeing you end with “perhaps a better is is to do this”..

        it’s Govt is bad.. I told you so.. over and out!

      • “I’m not blaming the governor for the outcome.”


        “Another Example of Good Intentions Gone Wrong”

        “Will McAuliffe rescind his ‘ban the box’ order? I’m not betting on it.”

        “The social engineer’s response to problems created by a law or regulation is to ‘fix’ the emergent problem by enacting more laws and regulations… thus creating new problems.”

        I don’t know why you think the people who come here are either stupid or illiterate. Your position is quite clear – McAuliffe is an ideological do-gooder who tried something that failed and the burden for the failure is on HIS shoulders not the racist employers who aren’t hiring people because their racism leads them to the conclusion that so many Black and Latino laborers are criminals they should just stop hiring them altogether.

        • I’m not blaming McAuliffe for the way employers responded to his initiative. I’m noting that he failed to recognize that employers didn’t respond in the way he anticipated.

          If the goal is to actually help minorities find jobs, as opposed to express good intentions or vent moral outrage, then the governor needs to try something else. But actually doing something that works isn’t nearly as satisfying as venting moral outrage, is it, LOTFL?

          • LifeOnTheFallLine

            The ban the box initiative wasn’t about venting moral outrage it was about acknowledging that it’s wrong to exile people to prison to pay for their crimes then exile them from job opportunities after they pay for their crimes. It was an attempt to address an actual injustice that racist employers responded to with more injustice, but that response doesn’t matter enough for you to write an entire post about how disgusted you are with employers, just how incompetent do-gooding ideologues within the government are.

            But keep pretending it’s anyone other than you venting moral outrage, it’s a really compelling and believable line of argumentation!

          • Wow, you are one bitter guy.

        • “Racist employers” (and most are not, as you very well know) are who they are; they are not elected, but reflect and respond to the world around them. McAuliffe is the Governor, whom we elected to make that world a better place (in his own partizan way, and undoubtedly with a little partizan pillaging along the way). LOFL, I think it’s quite appropriate to blame him for the unintended consequences of his executive order and leave out of it the employers who respond to it by even more drastic, indeed racist, filters of their own.

          What I will grant you is that ‘ban the box’ might be worth this cost if it advanced the cause of racism awareness — but what evidence is there of that?

          • LifeOnTheFallLine

            Blame everybody but the racists…what a distressingly familiar line of reasoning.

          • Reed Fawell 3rd

            This was a predictable outcome. I suspect the job loss is far higher than the numbers quoted in the study in many low skill sectors of the economy. It has nothing to do with racism. It has to do with the operating efficiency of many businesses and other employers, who must control key risks to serve their clients, and to best assure that one can honor ones contractual obligations so as to keep customers and make a profit so as to stay in business rather than incur chronic loses, screw up jobs, hurt other employees, and go out of business in the process of accepting unreasonable risks forced on one by ones government. The cost of blindly hiring unreliable, erratic and potentially dangerous employees can be astronomical. It can put employers out of business, particularly so, for example, in highly competitive bid situations, like construction. That is why drug testing is considered key for many.

            Far more serious efforts need be made by all those concerned to help these folks get and keep good jobs, but the “blind, blanket, and stroke of the pen” feel good legislation solves nothing, while it causes far greater harm to all concerned.

            Jennifer Doleac, the assistant professor of public policy and economics at the University of Virginia, has done a great public service. Hopefully it will lead to much needed real solutions.

  3. good to see you back LOFL!!!


    • I appreciate that, but I’m not really back because this blog still traffics far too much in racist thinking and gives far too much latitude to racist commenters for me to actively contribute to its success. I still check in from time to time to see what the Virginia right is thinking, and in this case comment because of the race issue involved, but as long as there are front page posts saying “Here’s a graph showing how test scores drop for people and communities at the lower end of the economic spectrum BUT MONEY CAN’T SOLVE THE PROBLEM” and talking about “Petersburg is the NEW DETROIT” while conveniently ignoring places like San Bernardino, CA; Jefferson County, AL and Central Falls, RI and act like it’s not race baiting nonsense then I’m good.

      • I get it. Right wing/conservative thinking = racist. Did you get mugged by Black Lives Matter? Do you not understand that your post is hardly the way to begin a meaningful conversation. Then again, perhaps that’s not your intent in the first place. Maybe some supporting evidence might help your tirade.

        I’ve learned over the years that when someone yells “racist”, in only the broadest terms and without a lick of supporting evidence, it’s better to turn a deaf ear.

  4. >>Individuals and private entities can quickly alter their behavior to adjust to reality; government adjusts much more slowly, if at all.>>

    This is a concept that LarryG is constitutionally incapable of acknowledging.

  5. It might be added CJD, that when they do alter behavior they probably have introduced even more complexity into whatever the process is/was … with apologies for mixed tenses .

    Please someone help me to remember when a government actually simplified a process or regulation.

    • I have something simple.. The GPS location on your phone.

      Now how simple is that? You don’t even know it’s doing it!!

      yet.. that phone is “talking” to a constellation of US Govt satellites and getting a precise location fix of where you are which is then used for GOOGLE Map navigation or UBER or 911… and a host of other “apps”.

      now how simple is that?

  6. Like those 50+ year old zoning codes in Henrico and Spotsylvania. Certainly, any re-do of these will be a simplification.

  7. yeah.. I still don’t think Bacon is really interested in offering better ways to achieve the intended goal as he is in counting coup …

  8. Of the 24 states which have passed the “ban the box” initiative only 9 apply the rule to private employers. The remaining 15 (including Virginia) only apply to state agencies. Even then, certain agencies like the state police are exempt.

    It’s hard to understand Dolec’s methodology. In the Daily Progress article it said, “After measuring employment rates in areas around the country that had adopted the policy, and incorporating controls for various other aspects that could affect those rates, Doleac and Hansen found their hypothesis to be correct for young, low-skilled black and Hispanic men.”

    How can you look at overall employment rates in states, cities and towns that have adopted ban the box policies for state or local government agencies and find statistically significant causality? Charlottesville is mentioned in the article. Charlottesville bans the box on initial job application forms for certain city agencies. Can you really look at overall employment trends in Charlottesville and ascribe any differences from the national mean to a checkbox on an initial employment form?

    I assume that Dolec performed her analysis correctly. I just wish that the article would have provided more details on the methodology. For example, I would liked to have seen the results of her analysis in those states where ban the box is applied to both private and public employers separate from the states where the policy only applies to public employers.

    The assumption of racism from some commentators is also questionable. Those who read the underlying article would have found this, “Doleac and Hansen’s findings are not without some bright spots. The study found that employers were more likely to avoid younger men, but not necessarily older men with convictions or incarcerations. Older black men, for example, were more likely to get jobs in states with ban the box.”

    My suspicion is that initial employment forms usually require the applicant to document his or her last 5 – 10 years of employment history. Absent the checkbox a gap in the timeline may be enough for prospective employers to discard the application under the assumption that the gap might represent a period of incarceration. Since African Americans and Hispanics tend to be incarcerated (and unemployed) at a higher percentage than their white peers there are more timeline gaps. Older candidates would likely have been incarcerated prior to the recent work timeline. There would be no gap. In those cases older black men were more likely to be employed since those who had been incarcerated no longer have to check that checkbox.

    I find the whole idea of trying to temporarily hide conviction information from employers questionable. The sad reality is that released prisoners have a very high rate of recidivism. According to the Bureau of Justice Statistics, ” Overall, 67.8% of the 404,638 state prisoners released in 2005 in 30 states were arrested within 3 years of release, and 76.6% were arrested within 5 years of release”. While better employment opportunities for former inmates would help reduce recidivism, it’s hard to imagine that those engaged in criminal lifestyles, as a group, will ever reach the arrest rates for the population at large.

    The only answer is to work with the prisoner while they are incarcerated to make that person a more attractive prospective employee. Education, job skills and even visible tattoo removal will do more to improve a former inmate’s chances of employment than regulating checkboxes.

    • There are a number of effective groups hard at work out there trying to help mediate that recidivist problem.

      One of those groups that I have taken a special interest in is Prison Fellowship. You can find it at

    • Good Job DON! You’ve nailed it on all counts!

      as you said – you’d want to keep the data to the group that the premise is about -and then you’d need to compare to a similar demographic group that does have the box.

      and yes.. I agree with point about the misguided rule that almost surely leads to just say no to low level entry jobs that they cannot afford to order their own background check on..

      but I will also point out that felons also exist in demographics that some folks don’t think do – not all young folks with the color and look that you suspect are felons.. some nice looking lighter skin folks are felons also.

      so if the premise of the study includes unemployment demographics – you need to do that level of granularity also and compare again to a test group.

      there’s entirely too many of “jack-leg” studies these days that promise to deliver any answer to a question but instead deliver a biased premise.

      finally – yes asking for a work history from a young person to account for the few years they were around after high school … “nothing” or “I don’t recall” answers are probably much more informative than their looks and skin color.

Leave a Reply