Contains law: open with caution

By skepticlawyer
Affirmative action and its whiff — at least as spotted on Planet Janet — came up on Legal Eagle’s thread on the appointment of Justice Bell to the High Court. I’ve since had a couple of emails — and participated in a lengthy thread elsewhere — that discussed it. People are surprised when I say I’m opposed to any form of affirmative action, especially when they know me — you won’t find a stronger supporter of individual autonomy. If a woman passes the required physical benchmarks for the fire service or for combat roles in the military, then I think those barriers should be swept away. This support doesn’t extend to official ‘encouragement’ for minority/female participation in historically male or majority roles, however. My view is only partly for the reason LE outlines — that the appointee (especially if the position is a very prestigious one) can look like a charity case. My opposition is based on something far simpler: affirmative action doesn’t work.

First, though, some clarifications: not everything is what it appears to be. What looks like sexism in employment, say, often isn’t.

If we found, for example, that females applying for CEO (or other senior managerial roles) are being passed over in favour of males with similar experience and skills, that would be evidence of discrimination on the basis of gender. If, however, we find that the males being hired usually have more experience, then we have evidence of discrimination on the basis of experience. Obviously enough, this sort of discrimination is widely approved — indeed, anyone who has ever been an employer has undoubtedly used this particular ‘control’. This means a careful comparison of candidates needs to be made before any ‘call’ can be made.  It’s vitally important to establish whether there are appointment disparities between similar males and females. 

If we are still interested in gender inequality in employment, it’s worth looking for the thing that is causing, say, the experience disparity. In study after study, the single largest impact on female career progression and income is motherhood; the more children a woman has, the greater the statistical likelihood that she will not fulfill career potential to the same extent as an identically talented male. While there are statistical outliers (Quentin Bryce, Sarah Palin), they do not affect the overall data set very much. 

When the presence of children is controlled for, the difference between male and female salaries and promotion records all but disappears. That having children has such a disproportionate effect is extremely well documented. Even in egalitarian social democratic Finland (to take but one example), children are a bombshell, and doing without ensures parity within a few percent.

Sure, you may say, that means we need to make it really easy for women to get support if they have children. This means paid maternity leave, workplace creches and so on. However, before advocates of both these things put their legal boots on, it’s wise to remember that legislators need to work out which is more expensive: women dropping out of the workforce in fairly large numbers or paid maternity leave. If it turns out that paid maternity leave is cheaper (ceteris paribus) than the economic losses due to female departures from the workforce, then that’s a prima facie case in its favour. Unfortunately, this isn’t the end of the matter — lots of things need to be considered first before we can approach an ‘optimal’ policy. I’m thinking of things like the length of time a woman should stay at home to care for her child (longer is better for her child up to about 5 years, and shorter — but not below 12 months — is better for the employer).

Then there are a mass of complicated social policy and philosophical issues still to one side. Is it right to force the childless to pay for those with children? Is it even possible for the state to regulate employer/employee relations in this much detail? (Studies show it is easier with large firms and with government bodies, the former because they are richer and the latter because they don’t have to deal with market signals). Should we just accept that altruism is weak in the human animal and that women — absent the social controls that existed 50 years ago — will simply make less altruistic choices and have fewer children?

This is no easy thing, as legislation that doesn’t align with people’s self-interest is almost always doomed to fail. Economist Stephen Levitt calls it ‘do good’ laws, and his account of what can go wrong is instructive when we come to the Americans with Disabilities Act and employment opportunities for disabled people. In a large study, MIT’s Daron Acemoglu and Josh Angrist made the following melancholy finding:

[W]hen the A.D.A. was enacted in 1992, it led to a sharp drop in the employment of disabled workers. How could this be? Employers, concerned that they wouldn’t be able to discipline or fire disabled workers who happened to be incompetent, apparently avoided hiring them in the first place [The original paper is here].

Affirmative action is an attempt to find an easy short-cut through statistical and legal minefields like this. There is no easy short-cut; if there were we’d have found it already.

It’s also worth remembering that because it is a regulatory mechanism (ie someone has to enact a law that the rest of us are supposed to follow), it will generate costs all its own. Laws need to be policed, administered and reviewed. Too many lawyers (not to mention the laypeople they often govern) assume law is costless and cannot be gamed. Very, very occasionally that is the case. More often, it creates perverse incentives and moral hazard as people learn to play a game of legal ‘musical chairs’ in order to further their self-interest.  Affirmative action in the US has lead to relatively stupid people being admitted to prestigious American universities and incompetent people getting management jobs. It is one reason why Harvard and Yale are academically deadlocked with Oxford and Cambridge despite being much richer (there is a fairly straight-line correlation between wealth and improved educational performance, at least at the tertiary level. Nothing like a bit of spare cash for a few scanning electron microscopes). Until very recently, a statistically significant portion of their intake has been admitted with grades below the ‘white’ cut-off (and, for nearly 100 years, Jews have been the real losers). There is no affirmative action at Oxbridge. 

This kind of discrimination may seem worthy and reasonable and something rich Ivy League schools and top-tier law firms can live with in the name of repaying a social ‘debt’… until one considers the effect an incompetent lawyer, doctor or engineer will have on the world around them. As I’ve often said, law is a broadsword, not a scalpel — let it out of its box with caution. 


  1. dan
    Posted December 19, 2008 at 12:04 pm | Permalink

    I had missed the addendum to LE’s post, but a further point is that discussing the question of affirmative action when it comes to appointments to the court is a more complicated question again.

    Though I know governments don’t always approach things this way, the balance of court benches (particularly in an institution as important as the High Court) demands diversity in a whole range of areas.

    We are no longer taught and no longer believe that judges can decide matters purely objectively without any influence from their background and life experiences. So diversity of the court arguably improves the quality and effectiveness of the institution itself in adjudicating the matters that comes before it.

    The dramatic increase in female jurists across the system in the last 30 years has almost certainly improved the quality of the institution, not because the women were so superior in ability to the men who might otherwise have been appointed (though most of them are excellent as you would expect), but if for no other reason that it is more likely that justice is seen to be done.

    To suggest that trying to achieve diversity in the bench is about affirmative action is a misconception. When an otherwise suitably qualified non-Anglo Saxon (particularly migrant or aboriginal) judge becomes available I am sure that it will likewise be a prominent consideration, not as a token or for affirmative action, but because it is currently a serious potential impediment to the perception of the court by the public – something that is of inherent importance to the performance of its function.

    Sorry about the long comment, but some of the commentary around the traps has very much annoyed me for failing to appreciate the difference between appointing judges to a higher court and other areas where affirmative action might be seen to operate.

  2. conrad
    Posted December 20, 2008 at 4:59 am | Permalink

    I’m personally not completely against scholarships and the like (although not on huge scales) — but targeting them at universities is the wrong place, since most of the damage has been done by then — it’s like trying to fix a broken arm with a panadol. If people are concerned about equity across different racial or cultural groups, then they would be far better off targeting schools, and preferably the teachers and school infrastructure and not the individuals.

    The other problem with all this targeting is that no-one knows when to stop. In 1960, helping women over men in some ways was a good idea. But in 2008, people are still trying to do it, yet women now have better overall outcomes than men in almost all areas, excluding wage parity for doing the same job and overall wages. However, even that is somewhat misleading, since last time I checked, around two-thirds of our (and the US) graduates are now female. My bet is that in the next decade or two, women that work full time will have higher average wages than men who work full time (but still slightly lower wages in the same job). What does one do about equity then?

    I think there’s also a destructive element to the idea that all learning should be homogenized, which was one of the things that came with all of this ideology — it ignores potential differences between learning styles in boys and girls. If girls happen to learn some stereotypical thing better in one way (e.g., poetry) and boys happen to learn some stereotypical thing (e.g., mathematics) better in another way, then we would should be using those better ways with each group to get the best out of both of them, even if it leads to differences, rather than serving up something that gets average but similar performance from both groups. I’m sure this latter idea is one of the reasons that mathematics scores of boys has dropped across the years (I believe mainly through homogenization of scores, with the right tail geting killed off). An essentially identical argument applies for different cultural groups, although even suggesting that these days gets you labeled as a wacko.

  3. DeusExMacintosh
    Posted December 20, 2008 at 2:35 pm | Permalink

    Scholarships aren’t the same, Conrad, these are usually based on identity-exempt measures of ability. The problem with ‘affirmative action’ is that it accepts lower performance thresholds depending on racial or other profiling. There isn’t as much of a tradition of that in the UK or Australia, it seems to be a US thing. It’s why places like Oxford and Cambridge are criticised for being elitist and ‘favouring’ people from private schools.

  4. Posted December 20, 2008 at 3:06 pm | Permalink

    The other problem with all this targeting is that no-one knows when to stop.

    When I was at Uni I was the student rep on a selection panel for an academic position. The other four were academics, three in the field, one from outside; two men, two women.

    The position called for a postgrad degree.

    Affirmative action removed this requirement for one candidate (she’d had a child). Yet her mentor kept pushing affirmative action and implying that we were sexist if we didn’t give her the job despite the fact she hadn’t displayed the qualifications or even the potential to gain them in any way.

    She didn’t get the job (thanks to me). I was appalled by the behaviour of everyone on the panel but giving someone a job they simply can’t do because of their sex does nothing to help women in general. It only provides ammunition for anyone who wants to say women can’t do the work.

    Thing is stuff like this happens all the time. But affirmative action is so entrenched as righteous in some quarters that even mentioning these sorts of pitfalls brings all sorts of accusations of bigotry down on your head.

  5. Posted December 20, 2008 at 3:09 pm | Permalink

    Affirmative action, that is to say, giving any kind of preference on the basis of race or gender or whatever (assuming it is unrelated to the occupational requirement such as a male hero and a female heroine) is discrimination, straight up. Thus affirmative action in the US (set asides and quotas) is the most obvious form of institutionalised racism since the abolition of slavery.

    As for how well it works, check out the international survey by Thomas Sowell.

  6. Posted December 20, 2008 at 7:51 pm | Permalink

    Interesting comments all. Dan – I’m not sure I buy the argument that people decide things differently because of their gender or background. That assumes that someone’s base character is formed based largely or wholly on their gender or background, and not on any of the active choices the person may have made. It’s dangerously close to essentialism, and perhaps even discriminates against people who don’t fit their particular gender stereotype — a gay chap who’s very ‘manly’ for example, a woman who’s ueber rational (something of which I can legitimately be accused). I do agree about justice needing to be seen to be done, but that does tend to involve the actual decision-making process, not who is on the bench.

    Conrad: I think there’s a lot to be said for scholarships, but if they’re to be given out using taxpayers’ money, they should be squarely aimed at economic disadvantage (which is what Oxford — and I assume Cambridge — does). I know it sounds dreadfully Marxist, but it’s poverty that holds people back, and it’s poor people who don’t get places at elite universities. Sometimes race and gender intersects with poverty, but sometimes they don’t. Private scholarships, of course, can be directed at anyone. There are a couple at my college directed at people who went to a particular high school, for example.

    DEM: Australia is catching the disease as well. Conrad can probably tell you more, but there are a range of ‘special entry’ provisions at Australian universities directed mainly at Aboriginal people, and I know people who’ve gained entry with no high school pass. Of course, they often fail, which makes the problem even worse — their self-esteem is down around their shoelaces by the end of it. Adrien’s anecdote is an illustration of this process in action when it comes to employment.

    I actually agree with Rafe. I suspect affirmative action is contrary to the 14th Amendment (equal protection clause). Some US states have abolished it on these grounds (eg California, Nebraska), but usually in response to Citizen Initiated Referenda, not any action by the courts.

    LE: I don’t know how the fundamental problem can be fixed either, as whatever anyone does — even a private initiative — is going to need to be a superb example of institutional design in order to avoid the fate of the Americans with Disabilities Act. Getting the incentives right will be of paramount importance.

  7. conrad
    Posted December 21, 2008 at 5:14 am | Permalink

    “they should be squarely aimed at economic disadvantage”

    I agree with this (I think scholarships for poor kids that make the cut is fine), but I believe race-based scholarships are fairly common in the US. It’s also not clear to me what the dividing line is between public and private in many universities is, even in the US, and also what one does about “part” scholarships, where the university does a deal to supplement the amount given from a private benefactor (potentially as an in-kind contribution), hence locking it into providing part funding for such a thing.

    As for what goes on in terms of affirmative action in Australian universities — there’s no official collation, but it is definitely common, at many levels. Where I work, for example, we get some students let in even though I know of no documentation for it (generally low performing minorities as far as I can tell), and then they inevitably fail or do very poorly, since no-one helps them catch up (I believe this is common across universities). They’ll be the new cash cow if some ofthe stuff in the Bradley report is implemented.

    Also, at least at the staffing level, one reason affirmative action is not going on in the UK as much is that in recent times, everyone needs to compete for funding at a department level. This was in part because of the much hated RAE. I believe one of the smart things it did was that it locked funding into specific departments. Thus, if your head of department happen to employ his friends (very common in Australia) or the equity officer forced you to take a few minority candidates that were no good, it meant that someone else could lose their funding and job if your department happened to lose its ranking (and hence had its funding chopped). The same was true for all those anachronistic hirings (you got your PhD from Cambridge, so you must be good). It pretty much killed them off too. Thus, at least for the good departments, everyone has a vested interest in getting the best person. Hopefully the Australian equivalent targets funding in a similar way (versus to, say, a faculty or university).

  8. Posted December 21, 2008 at 8:44 am | Permalink

    More random comments. First off, I did some time in the NSW Anti-Discrimination Board researching discrimination and intellectual handicap. Interesting to observe the tension between the aim of rewarding merit without preduce on irrelevant grounds and the aim of promoting various interests to achieve varius outcomes regardless of merit. The field attracted zealots with agendas who made careers in the (then) recently politicised public service.

    Among the results, (in addition to those identified by Sowell) a question mark over the actual merit of any famale/ethnic/etc who achieved advancement, and simmering resentment among the groups (mostly white Anglo males) who observed that they were being punished for the sins of their fathers and grandfathers.

    Also, as Sowell noted, the immediate beneficiaries were not the struggling masses of the disadvantated but well-placed, well educated and politically connected members of those groups.

  9. dan
    Posted December 22, 2008 at 9:30 am | Permalink

    “I’m not sure I buy the argument that people decide things differently because of their gender or background. That assumes that someone’s base character is formed based largely or wholly on their gender or background, and not on any of the active choices the person may have made.”

    Yes, I tried to distinguish my point so as not to go there, but I must need to be clearer. First, I think that people *are* affected in their decision making processes and even outcomes by their background, but not necessarily in the same way. For example a woman hearing a gender disctrimination case may reflect on her own experiences, but two women doing the same reflecting might make different decisions. The point is not that people’s decisions are determined by their background or anything else, simply that the old jurisprudential idea of the entirely objective jurist, just like the entirely objective scientist is simply not accurate. So I subscribe to a general “mutt” theory of the court – if we allow that people’s decision making is going to be affected by their different influences (even if not in predictable ways) then the most robust decision making body is a diverse one, with a range of influences.

    However, at the moment I find a great deal of difficulty with well-meaning programs designed to improve opportunities for women, such as the requirement in government tenders to demonstrate that you conform to a particular bar briefing policy by which X% of your briefs go to women. I have no problems briefing women (I am one after all), but there aren’t many of them who practice in the area I want to brief and several of that small number are on our “never brief again” list (along with plenty of men).

  10. AJ
    Posted December 25, 2008 at 8:35 pm | Permalink

    I don’t think you can blame AA for the problems of the Ivy League. Legacy pledges are worse than AA (but lack the social stigma), but the big problem is simply the law of diminishing returns, I’ve been to both an Aus and a US university, and the US one was better, but not 15x better, like the price would suggest.

  11. Posted December 26, 2008 at 12:26 pm | Permalink

    Legacy pledges are worse than AA (but lack the social stigma), but the big problem is simply the law of diminishing returns

    Excellent point.

    It’s funny how the Left’s preoccupation with collective solutions leads them to ignore relics of the feudal era. Those on the Right who’d sweep them away are constrained by an alliance with those who wish to maintain feudal relics. Hence the crazy situation where I might be allowed into my father’s alma mater even tho’ what he did and what I’ve done are in completely different fields. I haven’t designed any dams or tunnels. I struggle with long division. 🙂 .

    Fortunately, if you read the link, that’s not possible:

    The research shows that if you’re the first generation to university, that’s the hardest barrier to break. Once somebody’s been to university, they think about university for their children and their children automatically look at it as an option.

    Says Prof Mary Bownes Vice-Principal for Widening Participation, Recruitment and Community Relations (that’s a lot for a business card) but she also says:

    Let’s face it, our aim is to maintain and improve on our position as one of the best universities in the UK. And in order to do that we need to have the best students. People accuse us of being elitist and I suppose in that sense we are.

    That’s how it’s done.

    Giving people who can’t hack it, a university education, or worse, shunting them thru for reasons other than their having earned it the only way you can, simply renders a degree less valuable. End of story.

    Prof Bownes is the first in her generation to attend university and it must make it harder. In my family it’s hard to not attend. It’s expected. And that makes it easier. There’s no reason to make it any easier than that.

    Interestingly the best students I knew were first generation undergrads.

  12. Posted December 26, 2008 at 8:21 pm | Permalink

    AJ: the second link on AA in the Ivies in my post deals with legacies (and notes the disproportionate effect on talented Jewish applicants). The writer even goes so far as to suggest that they were developed to keep Jews out of the Ivies, and quotes a damning piece of private correspondence from one of the 1920s Harvard Presidents to that effect.

    The first link deals with affirmative action directed — mainly, but not wholly — at African-Americans, and impacting (now) disproportionately on talented Indian-Americans. It’s all of a piece to me.

    Oxford’s probably about an order of magnitude better than my Australian alma mater (so 10X), but that may be an effect of my decision to study law and philosophy, two fields where it enjoys global top spot. If you were a chemist, say, you’d be better off going to Cambridge.

    I’m a first generation to university – neither of my parents finished high school. You are right, it is a huge leap (and very hard to convince other people that it’s not just an expensive waste of time when you could be out earning money straight from school).

    I do think that ‘first gen’ people should focus on the professions, though, not the arts (which is the mistake I made). If you’re from so far outside the establishment that you don’t even know there is one, it’s almost impossible to manage your dealings with them (as I discovered). I should have listened to the guidance counsellor from the get-go — he said ‘law, medicine or engineering, leave the arts for rich poofters, no matter how much talent you have’. It was brutal, but ultimately sound advice.

  13. conrad
    Posted December 27, 2008 at 12:16 pm | Permalink

    “but the big problem is simply the law of diminishing returns”

    I’m not convinced by that. Now, it’s surely the case that for the average student, Oxford is not teaching, say, 10X what Deakin is, but you would really want to track outcomes over the long term. The reason for this is that many employers simply have no way to differentiate new graduates. Thus, if you are an employer and think the average Oxford graduate is, say, only 1.1 times as productive as the average Deakin graduate, then you will still select the Oxford graduate 100% of the time (we do the same where I work when selecting into the rather popular postgraduate program — we simple rule out graduates from some universities out because some proportion of their students look really good on paper but are not). Thus, there is sure to be quite a bit of non-linearity at this critical point in many people’s careers (there was a good article in the Guardian a while ago looking at outcomes for science graduates on a similar topic — where you go makes a huge difference). The other thing to consider is what would happen if you are a non-average student (say, excellent at one thing, awful at another). At least at the US universities, because they have very low staff-student ratios, they can actually help these students, whereas in Australia, they simply fall through the cracks (or perhaps gaping chasms is a better description) — thus there is a huge difference for some people.

  14. Posted December 27, 2008 at 2:17 pm | Permalink

    I do think that ‘first gen’ people should focus on the professions, though, not the arts

    Yeah. That tends to happen. It’s the get a trade approach to education. Studying non-vocatinal generalities like the humanities raises the suspicion that it’s a bludge.

    If you’re from so far outside the establishment that you don’t even know there is one, it’s almost impossible to manage your dealings with them

    Also true. A lot of Arts faculties aren’t serious. I got into Sydney Uni which, altho’ high status, had a dog’s breakfast of an arts course with a fragmented faculty. My experience in tutoring UQ philsophy students was similar. It seemed they were just required to pontificate on the latest trendy French dude. It’s all about creating a network of mutual admirers to collectively launch an assault at one of the plush positions in the Cultural Industries which operate on something quite similar to legacy pledges in this country.

    Griffith was much better as a freind of mine who studied classics at UQ told his teachers there. But now it’s got some status and (according to him) it’s slid in quality. The old GU mob I believe are in control of Arts at UQ now. I wish these things were more organized around teachers and teams than universities.

    leave the arts for rich poofters, no matter how much talent you have

    Ha ha ha 🙂

  15. Posted December 27, 2008 at 2:21 pm | Permalink

    I should add that my Dad who comes from a two centuries long line of engineers also thought studying film was a bludge. But this is a guy who couldn’t understand what graphic design was now matter how many times I explained it.

    I think they hack out the right side of the brain for engineers to make room for all the math.

Post a Comment

Your email is never published nor shared. Required fields are marked *