Why Erase Childhood?

In our society, adults must live with their records. We collect records on sport, contests, web-forums, marriage, school, jobs, crimes, debt, taxes, etc. Such records help others who want to interact with those adults, by helping them guess the consequences of such choices. Such records also help those who have good-looking records.

Of course, such records also hurt those with bad-looking records. Sometimes that hurt is unfair, as when a record looks bad due to a random event outside their control. But overall we judge it good to let people see records; we expect observers to usually take reasonable account of the possibility of noisy record signals.

For many kinds of records, we give the person who is the subject of the records the option to not reveal them. But we also let others draw inferences from such a lack of visible records. If a job applicant doesn’t show you a record of having graduated from college, you are allowed to infer that they probably didn’t go to college.

For children, however, we tend to go out of our way to prevent the collection and sharing of records. We often expunge childhood criminal records, and we make sure public schools don’t save or share records of grades and misconduct. Even though childhood behavior is often quite predictive of adult behavior. For example a larger literature (e.g., here, here) finds childhood misbehavior to be one of our best predictors of adult criminal behavior.

I don’t see an obvious rationale for this. The usual rationale for restricting kid behaviors is that the kids are irrational. But here we have a restriction on adults reacting to this person as an adult. The sorts of irrationalities someone displays as a kid are quite plausibly predictive of the irrationalities they might display as an adult. And I see no reason why adults should be especially irrational in interpreting such signs. We were all kids once, after all.

Yes kids who behave badly as kids will look worse as adults, and have worse life options and outcomes as a result. But we are mostly fine with this happening to adults due to their adult actions. What is so differently problematic about such things resulting from childhood actions?

GD Star Rating
loading...
Tagged as: , ,
Trackback URL:
  • Samuel Stringman

    Children are not considered responsible for many of their actions because they are not majority age. You cannot be culpable for things if you are not an adult yet.

    http://www.reddit.com/r/law/comments/2m3v1u/overcoming_bias_why_erase_childhood/

    • http://overcomingbias.com RobinHanson

      But we are not punishing the child – the child is not directly changed. It is the adult who has consequences.

      • Samuel Stringman

        Yes, but if a party was not responsible at the time of the crime, it should not go on a record as the party could not be culpable at the time. What’s the point of holding people accountable for things they are not culpable for?

      • Dan Browne

        Even for adults there are term limits on punishments. Take bankruptcy as an example. Declaring bankruptcy makes you a bad risk for some seven years, after which the records are expunged. Should all bankruptees be punished for the rest of their lives? Strikes me as somewhat harsh. Taking the other side of the argument, I’d argue that it would make sense to punish the adult for the sins of his/her childhood if it could be predictably inferred for the entire population that misbehaving children are ALL unreformable as adults. Otherwise it’s not fair.

      • http://overcomingbias.com RobinHanson

        Why don’t we expunge *all records for everyone after seven years? We can find out all sorts of things about what folks did ten years ago. Since the erased records are the exception, they need the exceptional justification.

      • Dan Browne

        I may be wrong on this but with respect to bankruptcy, the record is gone from the credit agencies after seven years. There’s also no legal requirement to maintain accounting records after some period of time (again I’m guessing, seven years). Other records (like criminal records) may hang around longer or even not be removed at all. Are you saying I’m wrong about the bankruptcy/credit agencies/government records? (And I’m happy to be proven wrong – just asking for information).

      • http://juridicalcoherence.blogspot.com/ Stephen Diamond

        You’re perhaps putting too much weight on “culpable.” Otherwise, I don’t understand the counter. Forget about who you’re “punishing”; you’re holding the child responsible.

      • arch1

        Robin,
        1) Would you therefore be OK with arresting, trying and imprisoning adults for crimes they committed at the age of (say) seven?
        2) The child is not directly changed, in the same sense that you would not be directly changed by learning that tomorrow would be your last day alive. I daresay that would be scant consolation as you lay awake tonight.

      • http://overcomingbias.com RobinHanson

        This post isn’t about deterrence of kid acts; it is about helping adults choose their associates well.

      • arch1

        Robin, I don’t think this is responsive to the point I was making, which is that certainty of future punishment can have large negative present consequences.

      • http://overcomingbias.com RobinHanson

        I’m ok with people caring about how their decisions today influence how their future selves decades hence are perceived. Kids included.

      • arch1

        I know. I was just reacting to your assertion that punishing the adult is not directly punishing the child. In any meangful sense, it *is*.

      • http://juridicalcoherence.blogspot.com/ Stephen Diamond

        I’m struggling to figure out why Robin doesn’t get this point. My current guess is that he construes the argument about age of majority as a purely legal point: the law says you can’t hold kids responsible in certain ways, but it doesn’t say anything about keeping records of their misdeeds.

        The point is rather (as I interpret it) that the same considerations that underpin the law of majority also justify the “erasure of childhood.”

      • mk

        It’s consistent with his other failures.

      • arch1

        That’s not real constructive mk.

  • ReS

    I might guess that most people view the child as a different ‘person’ from the adult, and thus consider it unfair to judge the adult based on the behaviour the child did, whereas people generally consider the adult a different person.

    I don’t think I buy such a response, but I would worry about letting childhood records out, because there may be knock-on effects for young offenders. It may be that, given job scarcity, the most rational thing to do after a young person committed a petty crime would be to continue in crime as chances of jobs afterwards would be slim. This is of course empirical, but I think it’s plausible.

    • http://overcomingbias.com RobinHanson

      Doesn’t the same argument say you shouldn’t turn an adult into a convict?

      • ReS

        I suppose the argument rests on the assumption that an adult will be more likely to offend in future if they have committed a crime as an adult, rather than as a child. I think it’s a factor to consider when thinking about turning adults into convicts, though.

      • Ken Arromdee

        I would make an even stronger statement that this is one of the big problems with our judicial system.

      • IMASBA

        An adult can be more easily deterred by knowing the consequences of their actions than a child can. And yes, for adults any but the most serious crimes records should also be expunged after a while, at least for employers and often this is already the legal reality.

      • Dan Browne

        An analogy is a person who has just started a new job. It’s expected that they will make some mistakes during the learning period (as long as they’re not material mistakes) and not be punished for them. Conversely, after say five years on the job, making rookie mistakes *ought to* be held against you because you *ought to* know better.

  • Scott Young

    I think there’s a lot of information about people (race, age, gender, childhood, medical history) that could be predictive in useful contexts. We make laws about using it at the expense of better decisions for a couple reasons:

    1. We worry about the broader societal effects of using such information. If we allow race to be a factor in employment, that may exacerbate current racial differences leading to negative societal externalities. This may even be a positive feedback, that minorities seek less education due to job market discrimination which reinforces the stereotype that they are less qualified.

    2. Certain information has a privacy value. Allowing the dissemination of that information may increase decision accuracy but at a personal privacy cost of each individual. My medical history may improve your ability to tell whether I’ll be a good worker slightly, but it may have a larger emotional cost on myself to expose a sexually contracted disease and public embarrassment. The same could be true of childhood misdeeds.

    3. The ethics of using certain information aren’t just about whether the information is useful, but what sort of incentives using that information would create. Would parents begin to cover up for their children, or control their childhood excessively to avoid a negative adult perception?

    • http://overcomingbias.com RobinHanson

      You don’t do much to say why these effect are stronger for kids than for adults. For example, yes parents might cover up kids, but friends and spouses might cover up for adults.

      • mk

        The intellectual dishonesty might be strong in this one.

    • anon

      > what sort of incentives using that information would create. Would parents … control their childhood excessively to avoid a negative adult perception?

      This is a good argument, actually. The well-known problems arising from wasteful signaling might be exacerbated in the case of childhood, because the de-facto decision makers (parents) do not act in the child’s best interest. This is quite comparable to arguments for barring children from working. It might look economically efficient at first glance, but the externalities are just too severe.

  • brendan_r

    Hypothetical: Some punk bullies you in high school. He steals from the cafeteria and vandalizes property. Are you anxious and morally obligated to give this dude a fresh start at your 10 year high school reunion?

    Yeah right. Why not? Because odds are he’s still gonna be a jerk.

    Sailer points this pattern out all the time. Many of the uncontroversial decisions people make at the local, personal, and community level, when generalized to the state and national level, look frighteningly reactionary (in the current intellectual environment).

    Anyway, publicizing juvenile records would have enormously disparate racial consequences. That alone is sufficient to explain widespread opposition.

    • http://overcomingbias.com RobinHanson

      Most of the info we can see about adults also has disparate racial impact.

      • brendan_r

        Look at civil service exams, which, like juvenile backgrounds, give info relevant to efficient decision making.

        Minorities and their activists used to like civil service exams because they precluded discrimination. Level playing field.

        Now they hate them.

        Civil exams are a clear example of fair, relevant and predictive info going from favored to suppressed entirely based on who benefits. Practical stuff.

        Maybe signaling explains the stronger preference for suppressing child records. But practical who/whom stuff matters too.

    • mk

      I’ve had just this experience and the reality was the opposite of your speculation.

    • Dan Browne

      Conversely, the kid who bullied me mercilessly in high school jumped into the middle of a bar brawl in my mid 20s and saved my butt when I was home visiting.

    • Thomas Bartscher

      My experience is counter to what you posit. Most of the people who where assholes when they were kids are actually pretty decent human beings now.

  • Yergen

    Whereas nearly all criminal adults were young offenders, most children that commit crimes are not going to be criminals as adults. Of course, different types of crime have different prediction value, but look at pot smoking, petty theft, fake id’s etc.

  • Unanimous

    I think society views childhood as a learning period in which making mistakes and pushing boundaries is a part of that process. A lesser degree of punishment is therefore applied, and this extends to removing negative records. I think the idea is to lessen the punishment on the child rather than the future adult that the child will become. This approach may not be fully thought through or entirely logical, but that happens sometimes.

    • John_Maxwell_IV

      Counterfactually speaking, if childhood records were public, parents would be harder on their kids and allow them to explore and experiment much less in order to keep their record clean. Consider a world where in order to look good for colleges, you didn’t only have to focus and get good grades in high school, but throughout your entire education. This seems like it would be a bad thing.

      I doubt it’s actually the reason why we erase kids’ records though. Let’s say someone is materially worse off due to a decision they made as a child, as a result of the policy Robin suggests. We would feel sympathy for this person, and we would think of foolish things we did during childhood that don’t have much to do with who we are as adults and relate to them. However, easy expunging of adult records means that adults have fewer incentives for good behavior. So no matter how much sympathy we feel for adults, we know that allowing expunging of their records means that they’ll behave worse, and adults behaving badly hurts society more than children behaving badly.

      In other words, adults can do more damage than children, so the reason we keep records of their behavior is so that they’ll be incentivized to behave well and not do damage.

    • Lord

      And growing as well as learning. Most would say they are different people as adults due to experience and maturation, knowledge and reasoning both growing over this period. We also treat those with mental problems and the demented differently even if we don’t expunge their records.

  • http://juridicalcoherence.blogspot.com/ Stephen Diamond

    Children bear less responsibility for their conduct because they don’t freely choose their circumstances.

  • Silent Cal

    Plausibly, childhood behavior is more strongly correlated with ‘suspect categories’ than adult behavior, so that erasing these records reduces undesirable discrimination. I think this explanation requires us to assume that employers and others who would look at childhood records use heuristics that do not conserve expected evidence; an example of such a heuristic would be ‘don’t hire people with criminal records’.

    • http://overcomingbias.com RobinHanson

      What suspect categories do you have in mind?

      • http://invariant.org/ Peter Gerdes

        Race, I suspect.

        The idea is that if you smoke weed or snort coke while black (especially as a kid) you are way way more likely to be arrested and if arrested for anything the police will not be sympathetic to the concerns of the parents about permanent records etc.. Moreover, you are seen as being dangerous rather than a harmless person engaging in youthful hijinks.

      • Silent Cal

        Racial, socioeconomic and cultural background are the obvious ones.

  • http://entitledtoanopinion.wordpress.com TGGP

    I was wondering the same thing when I read “Supercrunchers” on predicting criminality, as well as “When Brute Force Fails” on the inefficiency of having so many old people in prison when crime is a young man’s game.

  • Ben Albert Pace

    I will add a moral reason. Making childhood behaviour accountable sets up a market in good childhood records, meaning kid have to strongly consider their behaviour from birth onwards. This seems an unhealthy environment to rear new humans, and will probably cause a lot of stress (Take for example the situation with teenagers applying university. There’s a great deal of stress involved in making the perfect application, and getting the perfect grades.)

    • http://overcomingbias.com RobinHanson

      But we actually tend to lie to kids and tell them that kid behavior will go on their “permanent record”, when in fact it doesn’t. So we get the worst of both – we stress kids and we fail to have useful info as adults.

      • Jess Riedel

        Isn’t this mostly done by people who benefit in the short term from compliant kid behavior at possible kid future expense (teachers and parents) or people who are genuinely misinformed and are worried about kid future prospects (parents)?

      • http://overcomingbias.com RobinHanson

        Sure, but the fact that we allow it and don’t try hard to get ride of it suggests we aren’t focused here on reducing kid stress.

      • mk

        “we actually tend to lie”

        Speak for yourself.

  • Ken Arromdee

    “we expect observers to usually take reasonable account of the possibility of noisy record signals.”

    That’s the key to the problem.
    Imagine that the observer wants to do some concrete action based on the childhood record. An employer is a good example, though nothing is specific to employment here. The rational thing for him to do, other things being equal, is to always hire the person who has a better childhood record. The result is that a record which is very noisy and so only implies a 1% chance of criminality could mean a 100% less chance of getting a job–since employers will hire based on the best records and working their way down, if they stop before reaching the bottom, there will be an underclass whose disadvantage in the job market is all out of proportion to the worseness of their record.

    This is bad. And it applies to any sort of judgment based on a single unchangeable score that is only a weak predictor of bad things. (That is, it also applies to not allowing race, IQ, etc. in hiring.)

    • http://overcomingbias.com RobinHanson

      But that argument applies equally to all adult records as well.

      • Robert Koslover

        Yes indeed, but as adults they “should have known better.” I think do you have to set a standard of accountability at *some* age level, even if there may be no way to ever be 100% fair about it.

      • IMASBA

        Yes, you have to draw a line somewhere. People will always bitch about that, or write eloquent pieces on blogs to bitch about it, but in the end you have to have that line.

      • Ken Arromdee

        Hiring people based on race is a type of “adult record”; it’s a characteristic of adults which is correlated with criminality in a way that lets people make Bayseian decisions based on it. Presumably you don’t believe it is okay to hire people based on race. If so, doesn’t the argument for *that* apply to all adult records as well?

      • anon

        We ban race-based hiring because some folks have animosity against specific races, and we want to deter that kind of animosity. Does that argument apply to juvenile records in a way that wouldn’t be also true of adult ones?

      • Ken Arromdee

        I was asking why we ban race-based hiring that is analogous to juvenile-record-based-hiring: that is, there is a correlation between having a juvenile record/being of a particular race and criminality, and employers don’t want to hire potential criminals. This sort of hiring is not “based on animosity”, but based on a real statistical correlation.

        Why do we forbid this? If you can answer that, you have your answer of why we don’t let employers use juvenile records either.

    • anon

      > The rational thing for him to do, other things being equal

      But other things are not equal. Someone who is unfairly penalized because of small differences in records will probably compete by asking for lower wages, in order to make it in the employers’ interest to hire him. As employers thus learn more about how records relate to productivity, the disproportionate response will even out.

      Still, we should keep in mind that such an increase in info availability will necessarily cause increased inequality and perhaps even ‘underclasses’, as differences in productivity come to light and are reflected in diverging compensation. This can be addressed fairly easily by redistributing resources, so that less productive folks once again find themselves insured against their lower productivity.

      • http://invariant.org/ Peter Gerdes

        Unfortunately, wages at an employer are sticky (it creates office tension to hire multiple people for the same job at differing salaries in similar situations) and unless an employer chooses to specialize in hiring only people with a slightly worse juvie record you won’t get any situations where you can negotiate down your price like that.

        Worse, perverse incentives work against the company that chooses to hire only kids with JV records. Even when it would otherwise make sense to hire people with a slightly elevated criminal inclination grouping such people together makes the risk drastically larger. I mean suppose you have a 5 person accounting team that, jointly, has the power to embezzle vast amounts of money without detection. The probability that you will be embezzled from goes with the 5th power of the criminality percent of your workers. If a JV record doubles the small probability of criminality a company that chooses to hire only those with JV records has to reduce their pay ~ 6 times what they would reduce the pay of of a single hire with JV record on the team (things get worse if you worry about plans to get hired). This, plus the PR harm (do you want the company that as a policy only hires people with JV records to send over a repair man?) makes it difficult for anyone to hire them.

        Also asking for a lower salary has the perverse effect of suggesting to the employer that you only think you are worth that amount. To do this right the employer would have to be the one to counter with it’s offer but they can’t because of the stickiness problems.

      • anon

        No one’saying that a company should hire _only_ from folks with a record, or assign such folks to sensitive or customer-facing duties.

        > Also, asking for a lower salary has the perverse effect of suggesting to the employer that you only think you are worth that amount

        This is only true if you lowball _compared to the employer’s expectations_. The employer knows that you have a record, so she isn’t surprised by your requesting a lower salary – it doesn’t convey significant info.

  • blink

    Certainly we are throwing away a lot of useful data. I think we need to consider parents. They have a lot of control over kid’s environments and activities, so childhood records are much less one’s own than adult records. We already see too much wasteful competition as parents make their kids look good for school applications, but by erasing childhood records, we at least limit the benefits of this signaling.

    On the other hand, in a world of digital permanence, the point may be moot: There may be no escape from one’s childhood record.

  • jhertzli

    I’m starting to wonder about fetal records now…

  • GMHowe

    Several guesses:

    1. Childhood is a time of reduced moral culpability. Releasing juvenile records upon adulthood is punishing the child by reducing their future prospects.

    2. Most people tend to behave much more irresponsibly as juveniles so releasing such records would unfairly overshadow their adult records.

    3. Pressure from parents.

    4. Youth with criminal records have more potential than adults with criminal records. It may be more worthwhile to give a young person a second chance than an adult.

    5. We care more about children than adults.

  • Robert Koslover

    I agree with several of the folks commenting here. I’ve known some fine, upstanding, and undeniably-honorable (no kidding) adults who, in their youths, did less than honorable things. I know an excellent military officer, a superb father and a fine example to others, who in his youth had some run-ins (and deservedly so) with the law. I know an excellent physician who has saved many lives (and continues to do so), but who was actually booted out of public school for a while, due to disruptive pranks and (mild) vandalism. It really would be wrong to judge those people today by the children/teenagers they were. And I don’t think such old records could be made public without causing at least some embarrassment to them. Some misdeeds (some felonies, for example) are so serious that they should not be hidden. But we already have categories such as “juvenile, but tried as an adult” for things like that. But for the lesser things, let’s not get too carried away. Do you really, really, need to know if the respected 60 year-old pilot of the commercial aircraft that you are about to board was once caught shoplifting candy bars at age 11? Or, back at age 7, got into trouble at school for pushing a girl off a playground swing? Let’s allow honorable adults who have proven their worth to society to keep their less-honorable (just so long as they weren’t really, really, bad) childhoods private.

  • Cahokia

    Don’t conflate late adolescence with childhood.

    Despite recent controversies, 16 or 17 year old violent offenders are in the prime of their criminal careers.

  • IMASBA

    “Even though childhood behavior is often quite predictive of adult behavior.”

    That sounds like a politician, not a scientist. Yes, it is more predictive than a random guess but:

    1) it is less predictive than the same behavior during adulthood

    2) a large group does reform (but they might not be so willing or might simply be financially unable to reform if they knew one mistake you made as a youth can mean you will never get a job)

    3) most of these convictions are for pretty minor stuff and…

    4) minor stuff gets erased from adult criminal records all the time as well

    5) the record is horribly biased by the background of the parents (if you’re not from a poor neighborhood your parents can often prevent the thing from going to trial)

    6) the record is horribly biased by the rather low arrest rate among youths (most youths who do illegal things never get arrested for it) which brings us to…

    7) the reverse calculus: think about the probability that someone who does not become an adult criminal still committed illegal acts as a youth and didn’t get caught, in other words, use bayesian statistics and watch the predictive value of a juvee record decrease before your eyes

    8) juvee records are, on average, older and thus reflect older societal and judiciary attitudes

    9) you describe American and British attitudes, in continental Europe people are a lot less eager to share records with employers, insurance companies, etc… and more steps are taken to prevent the identities of non-convicted suspected to become public.

    • mk

      “That sounds like a politician, not a scientist.”

      Or, generally, someone who lacks basic logical and statistical competence. Even if true that “childhood behavior is often quite predictive of adult behavior” it could also be true that it often quite isn’t. It is precisely because of the sort of irrational inferences of people like Hanson that this information should be protected.

  • Army1987

    On reading the title of this post I guessed it’d be speculation about the evolutionary origins of childhood amnesia.

  • Viliam Búr

    I think we have this strong instinct to “protect the children”, which we follow even in situations where it doesn’t make sense. For example when a child bullies another child at school, our instincts prevent us from punishing the young bully, and thus we help the bullying to continue. People refuse to see bullying as serious, they frame it as a child’s play, because we have a strong taboo about calling any child evil, even if that child is actively trying to harm their classmate or drive them to suicide. You can still get an applause for saying: oh, but they are only a small child, we shouldn’t be so harsh on them; ignoring the fact that the child is hurting some other child.

    • http://invariant.org/ Peter Gerdes

      No, I don’t think that is the reason. The parents of the bullied child see it very very seriously. Even many school officials and the like do as well.

      Problem is that children of a certain age are just damn evil. They are experimenting with social power and part of that experimentation is learning to hurt others. As adults we only do this when we have something to gain but children have learning itself (no doubt evolutionarily knowing this made a difference) as justifying the effort and ill-will. Not every kid goes to the extent of bullying, better raised or nicer kids just play with ostracism and playing favorites, but enough are inclined to it that no amount of culling out bullies will really remove them (there will always be enough people left willing to take their place).

      Even adults very very upset about the level of bullying can’t effectively intervene. Bullying is a type of social harassment built up out of a pattern of treatment no instance of which is a big deal on it’s own. The bully might put on the show but bullying is enabled and causes harm through the peers who endorse or laugh at the victim’s pain. Adults are loyal enough, aware enough etc.. that they aren’t usually complicit in the harassment itself (some types of sexual harassment are exceptions…but much less so now).

      The authorities can catch the bully red handed but can’t punish him for more than they caught him doing which on it’s own is minor mischief. The more the bullied individual reports the bullying to adults who act or is singled out for protection the more the peers, experimenting with social power, see the game as more interesting and the target as more vulnerable showing weakness and the worse it gets. Even though the bully causes the action his behavior on it’s own really wouldn’t cause much distress. That’s what distinguishes a bully from a nemesis (who really shouldn’t be punished severely)…and how do you punish someone for “getting people to laugh” without just giving the punishing adult unfettered discretion?

      Worst of all is the fact that a no tolerance rule or stiff penalties on bullying tends to whip back on the victims. The bully has the rest of the class on his side when the victim fights back, usually overwhelmed with emotion, he has no lookouts, no one saying the other guy started it and has probably been deliberately pushed to initiating violence by verbal abuse. Yet, any really harsh policy about bullying ends up spinning in the wind when the bully or their parents can say “but see he hit me twice in front of a teacher and didn’t get in trouble and I just said something mean once and you’re expelling me!” Worse, even if a bully is scared off his proxies and earnest emulators will step in.

      No, it’s not a desire not to be harsh it’s the fact that bullying just isn’t an offense that adults can step in and correct.

      • Dan Browne

        Good post

    • Dan Browne

      You haven’t described the other side of the coin. The social interaction comprises of the bully bullying the bullied. Yes in many cases it is written off because the child could be just learning. In historical times, however, the solution was to encourage the bullied child to find a way to defend himself/herself and the outcome was that the bully learned that it wasn’t beneficial to bully. Today we discourage the bullied child from fighting back in addition to not punishing the bully and I believe *this* is another part of the reason why prevention of bullying fails.

  • Sieben

    Robin,

    I think we already DO have this sort of system. It’s just implicit rather than explicit. For example, my employer doesn’t know what kind of student I was in high school. But they can guess based on which college I graduated from.

    Just given my other background information as an adult, it is extremely unlikely that I was getting into lots of trouble as a child.

    Similarly, if I was a musician in some hoity toity ivy league orchestra, you might infer that I spent my childhood being yelled at to practice the violin.

    But I think that the value of knowing someone’s childhood record is extremely low if you already know their adult record.

    It’s also kind of hard to… *do* anything as a child that is really significant for an adult. Even if I get sent to juvenile for failing to attend school or disrespecting authority figures… well, those are things adults can do without being seriously punished. Conversely, if I’m the #1 LD debater in the state in high school, it’s big news. But if I’m an adult who still does LD debate it’s like a weird hobby and no one cares what my rank is.

    The stuff that is “big” and gets you noticed as a kid is really just trivial as an adult.

    • http://invariant.org/ Peter Gerdes

      Umm, assault with a deadly weapon. Conspiracy to defraud. Selling crack. Armed robbery etc.. etc..

      These all are, frequently, charged as juvenile offenses and would be a “big deal” as an adult.

      As I say above the ability of prosecutors to seek an adult trial for those they perceive to be especially culpable or dangerous means that there is an escape hatch that allows a judgement call to be made about whether it is really serious behavior (taking hostages at a bank with a real gun and apparent willingness to use it) or just bad youthful judgement…stumbling into the bank with the gun in your pocket and nervously asking the teller to hand you money after your friends told you to do it.

      • Dan Browne

        What about being attacked by a “bad kid” who is armed by a golf club and you happen to be playing baseball and use it in self defense but the emotional intensity and childhood lack of self-control leads to the “bad kid” having to take a trip to hospital? In an adult situation that would be “failure to show sufficient restraint” and would likely earn you a jail term. While I agree that there are certain behaviours that are likely red flags (especially if they are repeated and the kid shows no remorse), many (if not the majority) of children go through a learning process of self control. Holding them responsible thirty years fast forward is to me, far too harsh and I want no part of a world like that.

  • Jackson

    Do you believe:

    – Children have identical cognitive capability and understanding of long-term consequences as adults?

    – Children have the same level of control over their environment and choice of associations as adults?

    – Same level of influence when they are improperly charged with wrongdoing and appropriate recourse on par with adults?

    I would suggest this principle only makes sense if you answer yes to all three of these. The idea that children are equivalent to adults and we should collect data in the same fashion is likely to produce many false positives and be fundamentally cruel otherwise.

    I suspect the real societal aversion to this comes from the fact that nobody believes children are capable of operating on par with adults (or, if you believe that, I challenge you to loan your handgun collection to children or allow them to drive your motor vehicle to prove your beliefs).

  • http://invariant.org/ Peter Gerdes

    I think the difficulty of expunging the adult record explains much of the treatment.

    In general we would like it to be true that people with sufficient records of good behavior. I think most people would support the idea that if someone has gone for at least the length of their sentence without reoffending that information gets scrubbed from their record for purposes of employment in non-sensitive positions.
    But how would that work. As a child one interacts with a wholly different kind of largely governmental institution (schools) so it’s easy to leave employers in the dark about the particulars. As an adult you would have received adult prison time and have a giant hole in your record that you must explain.

    Worse, the information scrubbing selects just the wrong people and incentivizes the wrong behavior in adults. An adult who has been out of prison for some time will likely have a steady job. The scrubbing provision can’t make his current employer forget facts they had when they hired him so it primarily benefits those with less steady/good jobs and incentivizes reformed criminals to switch from situations they have found to work well by applying for work at a company that doesn’t know.

    A great deal of data gathering doesn’t happen for children or is required by law or practice to be done differently and it’s easy enough for background check organizations just not to collect juvenile convictions (by the time the info is requested they will be adults). It’s not so easy for an adult conviction. A single old record will let the secret out. Any prior application at the same company leaves the fact behind. Who knows where corporate mergers or headhunting firms have telegraphed the data in the mean time.

    Most importantly, as criminal trials of adults are public for important constitutional and transparency reasons those for children can be far less public (and social consensus prevents undermining it). Since the choice to prosecute as a child prohibits particularly severe punishments (and especially for minor infractions), the role of the parents in advising the child and different attitude of the judge and prosecutor make the relative benefits of a public trial less important. If the trial is fully a matter of public record it becomes virtually impossible to stop that information from being used by the public.

    The transition from child to adult provides a natural and socially accepted barrier to employer information requests and consultation of databases. The switch from a primarily schooling environment to a career prevents the problems of adverse selection and perverse incentives. Old data doesn’t follow you around.

    The fact that children accused of truly serious offenses are tried as an adult provides the escape hatch we need to avoid placing known monsters into sensitive positions. No predatory 16 year old sociopathic child rapist gets tried as a juvenile so we avoid having to define ‘sticky’ crimes or what jobs various criminals can and can’t hide their criminal background from.

    Lastly, and importantly, the length and nature of juvenile confinement itself allows for this erasure. Juvenile confinement is shorter, filled with much more study and support programs and doesn’t provide contact with studied, knowledgeable career criminals like prison. The mere fact of someone having served a 10 year term in a prison makes them a more capable, and dangerous potential employee.

  • brendan_r

    Imagine the US faces a war for its existence; or imagine it’s USA just before Vietnam explodes. A massive expedited draft looms. Who to draft; who to exclude; who to put in positions of authority. And we’re choosing amongst 18-24 year olds that we know little about.

    Childhood records would be invaluable. If you doubt that, think of the variety of your high school classmates: the clowns, thugs and studs. Think of which you want calling in airstrikes.

    You wouldn’t get any BS like, oh, yeah, Johnny likes to torture animals, but his home environment wasn’t any good, so make the kid an officer.

    Records predict outcomes. Don’t argue for suppression on the idea they don’t because Racism. Argue that the efficiency cost of suppression is worth it in terms of fairness, in your ethical system. But acknowledge the cost.

    • Dan Browne

      “Records predict outcomes”.
      No they do not.
      Many data points correlate to average outcomes.
      That’s *entirely* different than saying “Johny will be an evil psychopath because he beat up two kids when he was 10”.

  • mk

    “childhood behavior is often quite predictive of adult behavior”

    And it often isn’t. The gross intellectual incompetence of those who can’t grasp such simple facts is enough reason to protect these records from their irrational inferences. Where one is concerned about adult behavior, look at the adult behavior.

  • Avanti Shrikumar

    I might view it in terms of breaking cycles of poverty. Children from disadvantaged households are more likely to have done stupid things than children from privileged ones. An individual employer never has incentive to go for the candidate that looks worse. Since childhood behaviour is a little less predictive of adult behaviour than adult behaviour is of subsequent adult behaviour (because people change a lot between childhood and adulthood), hiding childhood records seems like a good place to introduce some noise into the decision-making process of the employers who, if they had more data, would never hire the higher-risk candidate. It’s not perfect, but I think that’s where it comes from.

  • http://juridicalcoherence.blogspot.com/ Stephen Diamond

    For children, however, we tend to go out of our way to prevent the collection and sharing of records. We often expunge childhood criminal records, and we make sure public schools don’t save or share records of grades and misconduct.

    I don’t think this is quite right; and putting it right strengthens your argument. Childhood isn’t erased; only events that detract from the child’s reputation. (Grades are misleading because the only realistic way not to share bad grades is to avoid showing the good. Still, let’s not overlook the obvious: high-school record determine college admissions.)

    Children’s achievements aren’t shrouded. (Some have been heard to boast of 7th grade test scores.)

    In this light, the so-called erasure of childhood (sic) is really more obviously irrational. (If children aren’t fully responsible for their misdeeds, they aren’t for their good deeds.)

    Apart from good reasons that do exist, the reason for shielding children, in this light, seems to express something like instinctive protectiveness.

    • http://overcomingbias.com RobinHanson

      Yes, that is a good point.

      • michael svehla

        What’s good about it?

    • IMASBA

      The “good” deeds of children are indeed not an airtight predictor of their later behavior either. But grades aren’t about good deeds, they are about intelligence and knowledge and those won’t disappear as a child matures.

  • mwotton

    One reason is that the pleasure+reward circuitry in the basal ganglia mature at about 13, but the pre-frontal cortex (responsible for evaluating consequences) doesn’t develop fully until about 23, which is one reason why teenagers do stupid things. In a real sense, they are brain-damaged and not yet able to be fully responsible for their actions.

    http://www.ncbi.nlm.nih.gov/pubmed/22081879

  • Thomas Bartscher

    The correlation between bad behavior in childhood and criminal activity of adults might very well be due to stigma. I know I’ve heard it very often in school that children are being told that they’ll never become functioning adults when they misbehave. Keeping childhood records would worsen the effect.