Blind recruiting is effective in eliminating discrimination, but the main bias in Australian Public Service recruitment is in favour of women and minorities, according to an experimental trial by the PM’s nudge unit.
There is discrimination in Australian Public Service recruitment, but it appears to generally favour women and candidates from minority groups, according to an experimental trial of blind recruiting with 2100 participants from 14 federal agencies.
It is commonly proposed that removing names and other information from job applications before shortlisting to hide traits like age, gender and ethnicity should result in fairer, more merit-based outcomes. The new study does not disprove this concept — but it does suggest that introducing blind recruiting across the APS would decrease rather than increase workforce diversity.
It was hoped that blind recruiting would increase the number of female and minority candidates. But the Behavioural Economics Team of the Australian Government (BETA) tested this hypothesis with a randomised controlled trial and found so-called “positive discrimination” in favour of these candidates is now so common in the APS that stripping personal details from CVs would actually hold them back.
BETA director Michael Hiscox writes:
“This is critically useful knowledge. It does not imply that the APS has solved the problem of gender equality at the executive levels and higher – or lack of diversity more generally – but it tells us that rather than putting the focus on bias in initial reviews of job applicants, it may be more valuable to direct attention to other stages of recruitment, including how positions are advertised, how interviews are conducted, and how hiring panels are selected and run.
“More attention can also be directed to processes that affect career trajectories, including performance reviews, evaluations for promotions, talent management and whether flexible working arrangements are available. Eliminating or mitigating problems in these areas will require innovative solutions and rigorous testing to discover what works.”
The randomised controlled trial found women were 2.9% more likely to be shortlisted and men were 3.2% less likely to get a foot in the door, compared to when personal information was removed. The boost increased to 8.6% for “minority females” and 5.8% for men who were also from a minority group.
Applications from Indigenous females were a massive 22.2% more likely to be shortlisted when these traits were visible to the person making the decision.
BETA also reports: “Interestingly, male reviewers displayed markedly more positive discrimination in favour of minority candidates than did female counterparts, and reviewers aged 40+ displayed much stronger affirmative action in favour for both women and minorities than did younger ones.”
Human resources staff demonstrated positive discrimination more than other public servants, picking females 9% more than their counterparts who couldn’t see the fake applicant’s gender, and shortlisting female minority candidates 41.4% more often than they did when assessing blind applications.
The positive bias varied considerably between agencies as well; some showed almost none while in the most extreme case, one agency was 55.4% more likely to shortlist men from minority groups, given the opportunity.
The researchers only stripped the mock applications of information about “gender, race and ethnic status” in this study but a whole range of other personal attributes could also be removed, if so desired:
“There are many potentially irrelevant characteristics that could be screened out from reviewers in order to remove biases. Besides gender, race or ethnic status, we might also consider any information on about age, health status or conditions, disability, sexual orientation, political views, and socioeconomic status (reflected for instance, by address and education background).
“De-identification could be implemented at the initial stage of the review process when applications are shortlisted and scored, but could also be extended to later stages (e.g., recruitment committee deliberations) – although it is obviously quite difficult (albeit not impossible) to de-identify candidates at the interview stage.”
Findings could stoke debate about merit
The report could easily stoke the simmering debate about whether workforce diversity policies undermine the merit principle, and could be taken as evidence that a truly level playing field is actually unwelcome in the APS.
Hiscox says his team’s results indicate a need for “caution” in moving towards blind recruitment, on the basis that APS leaders would want this “form of subtle affirmative action” to continue in their neck of the woods.
Blind recruiting supports recruitment processes where, in line with the merit principle, personal traits like gender or race are irrelevant to selecting the best person for the job. One might wonder why these traits are described as merely “potentially irrelevant” in the passage quoted above.
Critics of positive discrimination — who often come from the beneficiary groups themselves and argue their achievements are devalued by affirmative action — could easily attack this piece of advice as evidence that diversity policies only create an artificial appearance of fairness and equity that does not really exist in society.
One could interpret these new results — and the way they are presented — as evidence that a meritocratic process is less important in the APS than the need to rapidly boost workforce diversity figures. Blind recruiting is presented not as a way to make things fairer but as a tool that is only valuable where it can boost diversity.
According to the report:
“The overall implications of our study are that on average, across a broad range of APS agencies, introducing de-identification would have the unintended consequence of setting back efforts to promote more diversity at the senior management level in the public service.
“As things stand, senior public servants appear to be promoting diversity in the way they make decisions when selecting job candidates for shortlists during the initial stage of the recruitment process. This is not possible if applications are de-identified.”
The report notes there is only “limited and mixed” evidence about the effects of blind recruiting, internationally, and that personal attributes like gender and race become very relevant indeed, if encouraging positive discrimination is the aim:
“Several quasi-experimental studies in European countries have suggested de-identification could reduce bias in hiring processes in some contexts, but may have no impact in contexts in which no discrimination is present initially and, more perversely, may actually undermine efforts to promote diversity when employers adopt a positive bias in favour of women or minorities.”
The BETA report suggests that, depending on “the amount and direction of bias present in each agency” in the APS, there could be some specific areas where blind recruiting could work to increase diversity, in which case it would be welcome.
Should leaders bravely embrace affirmative action?
The study also leads to the question of how much affirmative action is too much. It is clear that in combination with the broader promotion of ideas like multiculturalism and gender equality over the years, workforce diversity policies are having the desired effect to some extent.
Is it the case that you can never have enough positive discrimination? Or does it risk becoming a fig leaf that devalues those members of disadvantaged groups who defy statistics through personal effort and are successful in their chosen careers?
The main counter-argument deployed by public sector leaders — articulated strongly by Australian Federal Police commissioner Andrew Colvin — is that individual assessments of merit have always been influenced by the unconscious bias, so policies like employment targets for certain groups — and presumably the positive discrimination they encourage — have the effect of balancing an already unfair situation.
There is a reasonable argument to be made that giving everyone a nominally equal shot is not enough to actually promote equity in outcomes. BETA’s report suggests that for leaders who believe this, blind recruiting might not be very helpful.
Both sides would probably agree that a genuinely level playing field would be the best situation, but creating one is extremely difficult and perhaps an idealistic fantasy. If one accepts the tables must be deliberately tilted to some degree, in order to find balance, how can anyone know how much pressure is the right amount?
Each group — Indigenous people, women, older people — also suffer different forms of discrimination in different situations for different reasons. If positive discrimination is believed to be the way forward, then public sector leaders need to bravely embrace it in more explicit forms so they can then get to work on all the details of exactly where, when and how to apply it.
Public sector diversity policies are also often underpinned by an aim to more closely reflect the general population, broken down in terms of the distribution of people in various identified groups. The idea is that if all things were equal, a large government workforce should mirror Australian society, but this seems more of an unproven assertion than a logical conclusion, or the self-evident truth it is often presented to be.
Like positive discrimination, this concept also seems to run against the merit principle. It certainly ignores the fact that all things are not equal. There is not necessarily a job in the APS that would suit every Australian, and the numbers of suitable candidates for any given job in the APS from each particular identified group is not necessarily in exact proportion to their distribution in the population.
But not so fast
The report notes a range of limitations on the study, including the possibility that people who volunteered to be involved were particularly keen on getting more women and minority candidates into the APS. Secondly, they point out that knowing their decisions were being studied could have affected the choices they made, and describe some efforts to mitigate this issue:
“Even though this was a familiar task for participants, it is possible that they behaved differently than they would in a real recruitment situation.
While we cannot control for the exercise being hypothetical, we were able to include design features that allowed us to mitigate reactivity and actually identify participants who might have guessed that the study was examining bias via their responses to some additional survey questions at the end of the exercise. When these few participants are excluded from the analysis, the results are unchanged. Our results would be further validated by a field trial using a real recruitment process to test the impact of de-identification of CVs on shortlisting.”
The study’s authors believe “it remains clear that more work needs to be done to address the problem of gender inequality” and point out that the shortlisting process at the start of recruitment is only one piece of the puzzle. But the main conclusion is clear: blind recruiting could well hinder rather than help agencies to hit their workforce diversity targets.