Redirected affirmative action: Entry and exit?

Abstract

Affirmative action is typically presumed to apply to recruitment but never lay-offs. Policies of affirmative action which only apply to recruitment are known as entry initiatives whilst exit initiatives involve lay-offs. In this paper, I challenge the status quo presumption against mixed exit initiatives which involve both lay-offs and preferential recruitment. Assuming that we have reasons to enact affirmative action initiatives whenever doing so promotes equality of opportunity, I strengthen the case for mixed exit initiatives by showing how, in at least some cases, the relevant initiatives will be likely to provide a greater reduction in inequalities of opportunity both within and across different birth cohorts. I consider various objections to exit initiatives as I construe them and argue that none of them are immediately decisive such that the status quo presumption is justified. I conclude by gesturing towards the plausibility of the initiatives discussed and other policies which merit further consideration.

Many advocates of affirmative action endorse what I call ex-post initiatives within the context of university admissions or hiring. In contrast to ex-ante policies like outreach programmes, additional educational resources, or ring-fenced funding opportunities, ex-post initiatives positively discriminate during or following competitive procedures with the aim of promoting equal opportunity for economically advantageous positions. ¹ These initiatives have been characterised as stronger forms of affirmative action and retroactively address inequality of opportunity through either quota or goal-based initiatives. ² Quota-based affirmative action involves reserving a number of places for applicants who belong to under-represented groups irrespective of how well these candidates do in a pairwise comparison with majority applicants. Contrastingly, goal-based affirmative action involves selectors ‘placing their thumbs on the scale’ with the aim of increasing the representation of minority groups, typically in order to achieve parity with the general population. ³ Both quota and goal-based affirmative action are matters which are fiercely contested politically and have recently been deemed unconstitutional by the Supreme Court of the USA in the context of university admissions. ⁴

In this paper, I work from the assumption that ex-post initiatives are justified as a means to redress inequality of opportunity. Assuming that either quota- or goal-based affirmative action can be justified in such a way and that we endorse the equal opportunity rationale in support of the relevant policies, I instead consider the following question: on whose shoulders should the burden of affirmative action fall? At the site of existing initiatives, affirmative action typically pertains only to recruitment rather than lay-offs and this is supported by the normative presumption in favour of entry initiatives. Since entry initiatives disrupt competitive procedures with either quotas or goals in order to influence the entry of women or minorities into advantageous positions, those who currently carry the burden of these policies tend to be young, majority candidates. ⁵ Conversely, exit initiatives shift the burden of the relevant policies to older, majority role-occupants who tend to be bigger beneficiaries of both vertical and horizontal opportunity-gaps.

In the first section, I begin by raising Judith Thomson's presumption against affirmative action policies which involve lay-offs as a stand-in for the extant position within both philosophical and public discourse that currently eschews exit initiatives in favour of entry schemes. I then proceed to differentiate exit from entry initiatives and suggest adopting a ‘mixed’ construal of the former that requires the implementation of either quotas or goals when rehiring for the open positions which exit initiatives generate. ⁶ In the second section, I strengthen the case for exit initiatives of this kind by elucidating the equal opportunity rationale for such policies and illustrate the ways in which they are able to promote equality of opportunity both within and across birth cohorts. Having elucidated pro tanto egalitarian support for mixed exit initiatives, I turn to consider seven different objections to the relevant policies in the third section. I argue that each of these objections cannot reasonably be taken to decide the case against exit initiatives without deciding the case against affirmative action tout court and so, the implementation of mixed exit policies remains plausible provided that one is committed to a similar justification in the parallel case of entry initiatives. I conclude by gesturing towards alternative policies which could be justified in a similar manner and contemplate whether my argument constitutes an inadvertent attempt at a kind of reductio ad absurdum. Since exit-based affirmative action is unachievable within the context of university admissions, I limit my discussion to decisions made in the context of hiring and selection for other advantageous positions.

Entry- v. exit-based affirmative action

Where a gap in opportunities between socially salient groups exists, ex-post initiatives are frequently suggested as a temporary policy to reduce persistent inequalities of opportunity. ⁷ Entry initiatives redress these inequalities by reducing the opportunity-gap between cohorts who are competing for particular positions within the labour market. Consider, for example, a society in which some minority group is subject to unjust disadvantageous differential treatment such that they now face worse economic opportunities relative to the majority. Whenever this inequality does not merely reflect differential choices which are the result of just socialisation and where affirmative action is the only viable means of addressing these inequalities, justice requires the adoption of entry-based affirmative action. ⁸ Both quotas and goals may be used in the context of entry initiatives and this constitutes the status quo in many respects. However, as Judith Thomson (2002: 50) makes clear,

it isn’t only the young white male applicant for a university job who has benefitted from the exclusion of blacks and women; the older white male, now comfortably tenured, also benefitted, and many defenders of preferential hiring feel that he should be asked to share the costs. Well, presumably we can’t demand that he give up his job, or share it.

It is this presumption, familiar to both philosophical and public discourse, which I wish to challenge. In doing so, the first difficulty lies in showing that exit-based affirmative action may be preferable from the perspective of equal opportunity. The second lies in demonstrating that there are no immediately decisive objections to exit initiatives such that this presumption is necessarily justified all things considered. In what follows, I address both of these challenges in turn and consider the pro tanto case for exit initiatives. Before doing so, I briefly clarify the distinction between entry- and exit-based affirmative action that I have in mind.

Both entry and exit policies constitute kinds of ex-post initiatives, and on my conceptualisation, our implementation of the latter does not have to be mutually exclusive to our implementation of the former. In the case of entry initiatives, affirmative action pertains to recruitment and typically aims to equalise opportunities amongst cohorts competing within the labour market for specific positions. In the case of exit initiatives, affirmative action involves lay-offs and aims to equalise opportunities between successive and senior cohorts respectively. ⁹ To see how exit and entry initiatives might be combined, first consider a case in which, for reasons of inter and intragenerational fairness, an institution decides to implement exit initiatives. In this scenario, senior role-occupants who make up the majority may be subject to involuntary lay-offs but this says nothing of the means through which these unoccupied roles should be filled.

The first and simplest option is to enforce an exit initiative of this kind before assessing prospective candidates on the basis of their institutionally determined merits without the use of either quotas or goals. In this case, the burden of ex-post affirmative action shifts entirely from young, majority candidates provided that entry initiatives are abandoned. ¹⁰ It does, however, leave open the possibility that these more senior roles are filled by senior, majority candidates once more and so may do little (or nothing at all) to redress the opportunity-gap between majority–minority candidates within senior cohorts or across different cohorts. The second and more egalitarian policy is to enforce an exit initiative of this kind and implement a complementary entry initiative when rehiring for the position rendered open by the senior role-occupant's exit. Since it is preferable from the perspective of equality of opportunity, I have mixed policies of the second kind in mind when referring to exit initiatives and so, exit initiatives, as I construe them, will involve some form of replacement. The implementation of these policies, however, could look quite different depending on the sector's demographic make-up, whether the institution hires external candidates or promotes internally, and the positions which complementary entry initiatives are applied to.

If, for example, complementary entry initiatives are applied to the positions vacated by senior role-occupants and institutions have to hire from an external pool in order to satisfy their requirements, then the burden of ex-post affirmative action will shift entirely from junior candidates within the relevant field or institution provided that the corresponding junior positions are unaffected. ¹¹ Alternatively, the burden of ex-post affirmative action might be shared more equally between different cohorts in the event that senior lay-offs trigger a series of internal promotions of candidates who satisfy the relevant quotas or goals. In this scenario, the exit of a senior role-occupant may instead lead to the vacancy of a more junior position given the way in which the position vacated is filled internally. Whether we think it is appropriate to apply entry initiatives to more junior positions in cases like this will depend upon the demographics of the sector which the institution is operating within and persistence of the injustice which affirmative action addresses. Clearly then, there are lots of ways in which the enforcement of exit initiatives could play out, but in each of the scenarios presented, the employer's implementation of the relevant policies leads to the creation of an opportunity for a position which would not otherwise exist at that moment in time.

Before proceeding to elucidate the equal opportunity rationale for initiatives of this kind, it is necessary to briefly pause here in order to address some potential confusion regarding my construal of exit initiatives as a mixed policy and an objection this conceptualisation might give rise to. It might be argued, so this objection goes, that by bundling entry and exit initiatives together in the way described, I am biasing my analysis in such a way so that exit initiatives will obviously better satisfy the equal opportunity rationale. As a counterpoint, it might therefore be suggested that a more stringent test for exit initiatives should involve avoiding the implementation of entry initiatives altogether. I think, however, that this worry is mistaken for a number of reasons.

Firstly, it seems that this objection risks misconstruing what it is for a policy of a certain kind to constitute ‘affirmative action’ to begin with. Admittedly, there is no unanimous agreement on a precise definition for the collection of policies which fall within the umbrella of affirmative action. But there is broad convergence on the thought that we should expect the relevant policies to increase the participation of women or minorities in specific fields through either ex-ante or ex-post initiatives, and that these policies redress under-representation which is thought to result from some form of injustice. ¹² For the reasons stated above, implementing an exit initiative without any complementary use of goals or quotas when rehiring for the position may merely constitute an arbitrary dismissal, as opposed to a particular kind of affirmative action which involves lay-offs. ¹³ On this construal of exit initiatives, there may be no reasonable expectation that the relevant policies will actually increase the representation of the relevant minorities. I think that this expectation constitutes an important criterion for understanding what affirmative action is or at least ought to be, and so helps us identify the relevant policies which fall within its umbrella. ¹⁴

Secondly, and returning to my reference to Thomson's presumption against exit initiatives as a stand-in for the philosophical and political status quo, it can be said without exaggeration that this presumption is one which disfavours any kind of scheme which involves lay-offs or other policies which I discuss below. It is not, I take it, the case that Thomson and others only take issue with what we might consider to be ‘pure’ exit initiatives, but the idea that affirmative action can permissibly involve (let alone require) lay-offs simpliciter. To challenge this presumption, it is therefore unnecessary to elucidate support for less egalitarian pure exit initiatives, as mixed policies of the kind that I have in mind will also be precluded by extant philosophical and societal norms.

Finally, the fact that a mixed policy of the kind which I have in mind might better satisfy the equal opportunity rationale for affirmative action in certain cases hardly strikes me as a reason to rule out further consideration of this construal of exit initiatives. On the contrary, if one supports affirmative action because the relevant policies promote equality of opportunity, then one ipso facto has reason to give due consideration to all of the relevant policies, especially when a particular policy is likely to better satisfy one's chosen justification in certain cases. That said, it might be the case that there are distinct moral or non-moral reasons which count against exit initiatives of the kind which I have in mind and that these reasons will settle the case against exit initiatives once they are given sufficient weight. I suspect that many readers sympathetic to Thomson's presumption might share this thought and so will consider the relevant policies to be unpalatable for one or more of these reasons. I will come to review these potentially decisive objections in the third section, but for now, I turn to further elucidate the equal opportunity rationale for exit initiatives as I construe them.

The equal opportunity rationale for exit initiatives

Since equality of opportunity is a forward-looking justification for affirmative action, it will not suffice to say that exit initiatives are justified by the thought that senior role-occupants should compensate the victims of historic wrongdoing by exiting their positions even if this line of argumentation looks convincing to those who are sympathetic to a compensatory rationale. ¹⁵ Instead, one must show that, in at least some cases, appealing to equal opportunity will provide as much, or more, support in favour of exit initiatives in order to challenge the status quo presumption. To wit, if you are an advocate of affirmative action as a means to promote equal opportunity in non-ideal circumstances, then you should, ceteris paribus, have reason to support initiatives which are likely to provide an equal, or greater reduction in inequality of opportunity between socially salient groups.

To illustrate the support for exit initiatives on these grounds, first consider an individual who you would deem to be a member of the majority who is advantaged in the relevant sense (e.g. a white European, heterosexual, able-bodied, male professor from a wealthy family). Since it is possible for a socially salient group to experience advantages even if some of its members face comparable disadvantages to the out-group, consider this individual to be a hypothetical ‘statistical person’ for the purposes of argumentation, given that proponents of affirmative action typically target the reduction of inequality between groups as a matter of primacy. ¹⁶ Moreover, since young Asian Americans remonstrate that they currently share the burdens of affirmative action in the USA, I hereafter distinguish between advantaged and disadvantaged groups in place of the majority–minority distinction, where ‘(dis)advantage’ refers to the global (dis)advantage one is statistically like to face with reference to opportunities for advantageous positions qua membership of a socially salient group. ¹⁷

Here, it is worth distinguishing between equality of opportunity's formal and substantive components. ¹⁸ Formal equality of opportunity requires advantageous positions to be open to all citizens and selected for on the basis of a candidate's institutionally determined merits. ¹⁹ This requires the removal of formal barriers through anti-discrimination laws and the enforcement of impartial procedures. Substantive opportunity is typically understood as a principle which aims to neutralise or mitigate the effects that one's starting place has on the means to attain the qualifications necessary to do well within these competitions and attain the goods associated with them. ²⁰ Precisely what this requires is subject to much contention. For expediency, I follow others in understanding the requirement as one which gives us reason to move towards a state in which people are more likely to enjoy equal prospects of economic success given the same native talent and ambition. ²¹ That said, I only assume here that some contemporary theory of equality of opportunity is compelling as a requirement of distributive justice and that a commitment to it generates moral reasons to promote substantive equality of opportunity in the sense described. What I have in mind here is a principle of justice that regulates the distribution of opportunities for advantageous positions, and so one that regulates one of the most important means to realising income, wealth, and other immaterial goods like respect or fulfilment. ²²

With this relatively loose conception of equal opportunity in view, we can say that affirmative action is typically justified by considerations of substantive opportunity, since the formal component is relatively undemanding and satisfied more regularly in most liberal democracies. Despite this, significant complaints of substantive opportunity persist, and affirmative action is a viable way of redressing these inequalities. Consider, for example, the finding (Schwitzgebel and Jennings, 2017: 86) that women constitute roughly half of the population but were reported in 2017 to hold 25% of US faculty positions in Philosophy Gourmet Report rated departments. Supposing that such discrepancies are the consequence of some combination of unjust socialisation, discrimination, and unequal access, we might conclude that women enjoy fewer opportunities in this domain despite the march towards formal equality (cf. Kallens et al, 2022: 672). ²³ It is a conclusion of this sort which leads many to endorse ex-ante and ex-post initiatives in order to redress the substantive inequalities of opportunity which socially salient groups face.

However, just as inequalities between genders or ethnic groups generate concerns of distributive injustice to which affirmative action responds, various inequalities between different generations present similar problems which merit further attention. Although age-group justice concerns the distribution of goods at stages of life which we all move through, birth cohort justice concerns the distribution of goods between successive cohorts (Bidadanure, 2021: 32). Unlike age-groups, cohorts are fixed at specific moments in time and are usually delineated with reference to the generation in which one is born. Baby-boomers, for example, constitute the cohort of people who were born between the Second World War and the 1960s, whereas millennials comprise the cohort born between 1981 and 1996. With cohortal justice in view, now consider how the presence of what I call vertical or horizontal opportunity-gaps focuses our attention towards the inequalities which arise between, or within, different cohorts.

Assuming a diachronic ‘complete lives’ view of equal opportunity, horizontal opportunity-gaps refer to the substantive inequalities of opportunity that exist within a particular birth cohort. If advantaged millennials, for example, are statistically likely to enjoy greater opportunities than their disadvantaged counterparts over the course of their lives, then these millennials are the beneficiaries of a horizontal opportunity-gap. Vertical opportunity-gaps refer to the substantive inequalities of opportunity that exist between successive birth cohorts. If advantaged baby-boomers enjoy greater opportunities than the birth cohorts which succeed them, then these baby-boomers are the beneficiaries of a vertical opportunity-gap. Unlike synchronic accounts of distributive justice which restrict our concern to a particular timeslice, a diachronic view of equality of opportunity directs our focus towards the set of opportunities that individuals are likely to enjoy over the course of their lives. Within the context of the labour market and affirmative action, diachronic assessments of opportunity therefore prompt us to evaluate the career and socioeconomic trajectories of a given cohort, rather than the specific opportunities they face at a given stage in their lives. As Juliana Bidadanure (2021: 24) motivates the concern, diachronic assessments of equality help ‘ensure that members of a given birth cohort do not end up worse off than the previous cohort through no fault of their own’. One might question, however, if it is really the case that younger cohorts are likely to face comparably worse substantive opportunities.

In support of the worry that this question raises, a sceptic might gesture towards the fact that later cohorts attend university in greater numbers, and have access to better technology, educational resources, and more knowledge. That said, it strikes me that advantaged baby-boomers are likely to have benefited from fewer affirmative action policies in the past and from entering the labour market at a younger age, with less debt and competition, at a time when states were further from the achievement of formal equality of opportunity. Contrastingly, disadvantaged millennials are likely beneficiaries of a greater number of contemporary affirmative action initiatives as well as the formalisation of fair selection practices, and so it may be reasonable to suggest that intragenerational fairness has improved over time. ²⁴ More concretely, junior cohorts are acknowledged to be subject to greater job insecurity, unequal benefit ratios, wage stagnation, unequal regard, and rising income inequality (Bidadanure, 2021: 1–4). They are also are much less likely to benefit from the security of home ownership and the accumulation of wealth which has followed decades of property value appreciation. ²⁵ Although these distinct intergenerational inequalities may be independently concerning, it is important to stress here that they are also objectionable because they are likely to leave younger cohorts worse-off with regards to the substantive opportunities they enjoy over the course of their lives. ²⁶

Putting the sociology of equality of opportunity aside, I think that we should draw three plausible conclusions from this discussion which provide pro tanto support for exit initiatives. Firstly, once we recognise birth cohorts as socially salient groups in which individuals experience inequalities qua group membership, one's cohort can be recognised as disadvantaged in the relevant sense and so affirmative action may target the reduction of vertical opportunity-gaps. Secondly, since the further back one goes, the less likely one is to find affirmative action initiatives or competitions which uphold formal equality of opportunity, advantaged senior cohorts are beneficiaries of bigger horizontal opportunity-gaps than their younger contemporaries. ²⁷ Thirdly, given all of the aforementioned intergenerational inequalities and relative prosperity that advantaged members of senior cohorts enjoy, these individuals are also beneficiaries of vertical opportunity-gaps and intergenerational inequality of opportunity. Consequently, exit initiatives of the kind discussed may be able to provide a greater reduction in inequality of opportunity by shifting or sharing the burden of affirmative action amongst those who are bigger beneficiaries of the inequalities which the relevant schemes are supposed to address. This does not imply that past beneficiaries bear special moral duties of compensation. Rather, it follows from the fact that because they enjoy significantly greater intra and intergenerational opportunities, they are justifiable targets of policies designed to promote equality of opportunity. A commitment to equality of opportunity will therefore generate reasons in favour of exit initiatives in cases where their implementation is likely to provide a greater reduction in either horizontal or vertical inequalities of opportunity.

At this juncture, however, one might be prompted to scrutinise the framing of the equal opportunity rationale provided here given my references to senior role-occupants who are statistically likely to be bigger beneficiaries of both vertical and horizontal opportunity-gaps. Indeed, when this claim is coupled with my expressed intention to investigate who should bear the burdens of affirmative action, the framing of my argument may raise doubts about whether the case I make for exit initiatives is truly motivated by a forward-looking conception of equality of opportunity. Given some of the rhetorical moves made here, it might instead be suggested that my argument is better understood as one which flows from a kind of ‘beneficiary pays’ rationale – the idea that those who have benefited the most from historic injustice should bear the largest burdens of affirmative action. Following this line of reasoning, one might object that I have misconstrued the normative support for exit initiatives (or my own argument at the very least) and that as a consequence, I will be unable to attend to the very different problems which a compensatory rationale of this sort might present. ²⁸

This worry, however, is predicated on a misreading of the argument developed here. Although I appeal to the fact that advantaged senior role-occupants tend to be beneficiaries of bigger vertical and horizontal opportunity-gaps, I only do so to reveal alternative targets of affirmative action policies which may be more effective because they are likely to narrow a bigger opportunity-gap. The injustice that I refer to here is not merely historically situated but persistent with respect to the distribution of opportunities which individuals face throughout their lives. My claim is not that these individuals owe a debt for historic advantages, but rather that their current socioeconomic standing makes them an appropriate target of policies which seek to equalise the distribution of intra and intergenerational opportunities. Indeed, as I alluded to above, members of junior cohorts are noted to be subject to unequal benefit ratios, wage stagnation, unequal regard, as well as greater job and housing insecurity. Comparatively speaking, these issues are only exacerbated when looking to disadvantaged members of junior cohorts and advantaged members of senior cohorts, since the former not only face many of the difficulties noted above, but face them in addition to other structural disadvantages which they experience at the intersection of other socially salient groups.

Recognising the effects of these socioeconomic circumstances on opportunities for the relevant goods helps shape the policies that we have moral reasons to endorse. The aim of affirmative action, however, under a forward-looking conception of equal opportunity, is not to compensate for these other wrongs or inequalities per se, but to implement schemes that are likely to mitigate the effects that these socioeconomic factors have on access to the goods which equality of opportunity regulates distribution for. ²⁹

The equal opportunity rationale for exit initiatives, as I have construed it, is not therefore to compensate for these inequalities by tinkering with policies which regulate access to positions, but to equalise opportunities over the course of different individuals’ lives by looking to the disadvantage they experience at the intersection of their cohort, gender, ethnicity, sexual orientation, and so on. Acknowledging that senior role-occupants are statistically likely to be beneficiaries of bigger opportunity-gaps does not require that we adopt a backward-looking justification for affirmative action any more than recognising that white millennials are likely to enjoy greater substantive opportunities than their black peers does. In both cases, the forward-looking rationale for affirmative action is the same, that implementing such schemes is likely to promote equality of opportunity overtime, but pinpointing the intersection of different disadvantages experienced over time and across different cohorts helps to reveal policies which we have as much, or more, reason to endorse.

In sum, though part of what I have said here hinges on the fact that senior role-occupants are the beneficiaries of greater opportunity-gaps, this acknowledgement does not imply the adoption of a backward-looking, beneficiary pays justification for exit initiatives. On the contrary, it serves to reinforce the forward-looking commitment to equality of opportunity by highlighting the persistent inequalities which policies of affirmative action are typically designed to redress. By focusing on how inequalities of opportunity manifest across different cohorts and intersect with other features of an individual's identity, we reveal where interventions can most effectively promote equal opportunity moving forwards and create normative space for initiatives which address intergenerational and other forms of injustice.

Redirected affirmative action?

The preceding discussion illuminates the equal opportunity rationale for exit initiatives. More, however, needs to be done in order to convincingly challenge the status quo presumption, since as I noted above, there may be decisive reasons to resist the implementation of exit initiatives in spite of the rationale elucidated above. I now turn to consider seven distinct objections to exit initiatives as I have construed them and strengthen the case for redirected affirmative action by rebutting them. Contrary to the presumption against exit initiatives, the following objections are not only surmountable but either beg the question in favour of the status quo or constitute reasons to reject ex-post affirmative action altogether. Supposing then, that one considers affirmative action to be justified insofar as it promotes equality of opportunity, one should be wary of too readily dismissing exit initiatives or reconsider their reasons for endorsing affirmative action tout court if they find the policy proposed unpalatable.

Reasonable expectations

One might object that a 64-year-old male professor who is advantaged in the relevant sense has a reasonable expectation that they will not be fired and that we should not introduce policies which contravene people's reasonable expectations. But as Lippert-Rasmussen (2020: 19) notes in his brief consideration of exit initiatives, this difference in expectations is largely ‘the result of the fact that affirmative action is rarely exit focused’. As such, this objection begs the question in favour of the expectations which the status quo cultivates and provides no meaningful support for the claim that these expectations ought to be morally privileged. Moreover, pressing the idea that one ought to give special weight to status quo expectations only raises the question as to why the initial move towards the deployment of various policies of affirmative action was justified at all given the parallel shift in expectations which accompanied it. If proponents of entry initiatives believe that it was justified to implement the corresponding policies at a time when individuals had reasonable expectations that they would not be subject to them, then it strikes me as odd for the same proponents to rule out exit initiatives on the basis that senior individuals do not expect their implementation. ³⁰

Feasibility

Putting aside the matter of whether or not exit initiatives risk contravening people's reasonable expectations, one might contemplate the potential infeasibility of exit initiatives and argue that justice does not require the enforcement of the infeasible. ³¹ Although the notion of feasibility relevant to political theory can be characterised in a number of ways, I wish to avoid any interminable debate about what the ‘feasible’ consists in here and so defer to David Miller's distinction between political and technical feasibility. By Miller's (2008: 46) lights, technical feasibility refers to ‘physical laws or rock bottom social or psychological laws’, whereas political feasibility refers to whether or not a proposal ‘can command sufficient political support to be adopted’. With this distinction in view, we can comfortably say that exit initiatives will not contravene any kind of technical feasibility constraint, since they are clearly consistent with what is technically possible. Exit initiatives might, however, be deemed to be politically infeasible for various reasons and this thought should not be surprising given the controversy which affirmative action faces more generally.

That said, whilst it is likely that exit initiatives would face significant pushback, I do not see why proponents of the equal opportunity rationale for affirmative action should take this fact to decide the case against exit initiatives. ³² If we think that justice requires equality of opportunity and that affirmative action is a technically feasible way of addressing the relevant inequalities in non-ideal circumstances, then we have reason to consider implementing policies like exit initiatives which address these injustices. The thought that these policies might be subject to criticism or controversy seems to constitute no more reason to discount our pursuit of them than it does in the case of entry initiatives. Here, one might be tempted to gesture towards the pushback that such initiatives may be likely to face and argue that our time is better spent supporting more politically feasible policies, or ones which are less likely to risk polarisation and division. However, exit initiatives are not only theoretically feasible but currently deployed by some institutions operating today. The University of Oxford, for instance, has, since 2011, instituted a pure exit policy which they refer to as the Employer Justified Retirement Age (EJRA). The EJRA mandates retirement by 70 at the highest grades of employment and is utilised to further intra and intergenerational fairness. ³³ Regarding the latter concern, I concede that exit initiatives are likely to be politically divisive, but I do not see how the relevant schemes differ from other policies of affirmative action in this respect and so it would be odd for proponents of these policies to reject exit initiatives on these grounds.

Efficiency

Assuming there are non-moral and moral reasons to promote efficiency, exit initiatives might be thought to produce undesirable and objectionable inefficiencies. For instance, one might worry whether those hired through complementary entry initiatives to fill the roles vacated by those exiting are liable to undermine efficiency, either because they are disadvantaged candidates from younger cohorts with less experience or because they are less qualified contemporaries. Whilst, in some cases, an interest in efficiency may be sufficient to defeat an employer's power to enforce exit initiatives (e.g. because the role-occupant is in one way or another irreplaceable), I see no reason as to why considerations of efficiency should necessarily rule out exit initiatives altogether and these considerations are rarely thought to prohibit other initiatives despite similar risks.

Discrimination

Avoiding any appeal to efficiency, feasibility, or reasonable expectations, one might object to exit initiatives on the grounds that they constitute a kind of age-based discrimination (Nussbaum, 2017: 60-62). Just as we recognise the injustice committed when particular ethnic groupsare subject to disadvantageous differential treatment, one might view exit initiatives as policies which give undue priority to the young whilst discriminating against the old. Although I am unable to provide a detailed analysis of competing theories of wrongful discrimination here, I believe that these worries can be assuaged with a brief rebuttal. Firstly, it is important to note that those affected by exit initiatives are not targeted because they belong to an older birth cohort, but because they are advantaged in the relevant sense. The fact that this is likely to intersect with one's age in some cases is a consequence of the fact that they are advantaged and so, being charitable, exit initiatives might constitute a kind of indirect discrimination.

Secondly, supposing that exit initiatives are indirectly discriminatory and wrongful, what does this say about entry initiatives which disproportionately affect those who are already unequally burdened by vulnerabilities in the labour market and other forms of intergenerational injustice? Either one accepts the force of this objection and forgoes ex-post initiatives altogether on similar grounds, or must point towards some salient normative difference. If anything, a salient difference between the cohorts discussed is that older cohorts are bigger beneficiaries of structural injustice with respect to gender, ethnicity, age, and opportunity. For those who think that structural or historic injustice is a necessary feature of wrongful discrimination, this point should be particularly compelling.

Pipelines

Another worry worth considering here is whether the point of affirmative action is to build better pipelines into advantageous professions and careers such the disadvantaged enjoy better prospects of success. It might be thought, however, that exit initiatives are unlikely to solve this problem given their focus on senior roles. ³⁴ In response, it should be stressed once more that the endorsement of exit initiatives does not require that we abandon entry initiatives altogether. In fact, we might endorse exit initiatives of the kind discussed whilst still applying entry initiatives in the context of university admissions, which plays an arguably bigger role in the pipeline to advantageous positions. Moreover, though an imperative to fix leaky pipelines is indeed persuasive, this paper has argued from the equal opportunity rationale for affirmative action. If one is moved to enact affirmative action policies in order to promote equality of opportunity, one will, ceteris paribus, have reasons to endorse exit initiatives in cases where they are likely to provide a greater reduction in the substantive inequalities of opportunities between different groups.

Overburdening

A further objection which merits attention appeals to the increased socioeconomic costs which may be associated with the implementation of exit initiatives. Since many states are struggling to maintain adequate social security schemes in the face of ageing populations and increased life expectancy, exit initiatives might be thought to worsen these challenges and overburden junior cohorts who will have to work harder and longer to fund the systems which support senior citizens. ³⁵ If so, then the implementation of exit initiatives would appear to impose a potentially significant cost on junior cohorts.

Although I believe that we should take this objection seriously, it strikes me that the size and impact of this cost will depend upon a number of factors, including the broader economic landscape and the way in which institutions implement exit initiatives. If combined with other policy measures, such as adjustments to pension structures and wealth or property taxes, then these initiatives could function as part of a larger strategy which addresses socioeconomic immobility and intergenerational inequality. However, even if such policies are unlikely to be implemented, exit initiatives may leave younger cohorts no worse-off than they are already destined to be. As things stand, the status quo arrangement does not distribute economic pressures evenly across generations; younger workers experience wage stagnation, job insecurity, unequal benefit ratios, high housing costs, and fewer opportunities for wealth accumulation. These factors already place a disproportionate strain on them, regardless of whether exit initiatives are implemented and so constitute something of an entrenched reality. If, however, exit initiatives exclusively target the most advantaged members of senior cohorts, then this may actually serve to ease rather than intensify these burdens, by allocating additional opportunities for economic advancement in a way that could mitigate some of the long-term economic disadvantages of intergenerational inequality.

Levelling-down

Finally, one might question whether the argument provided here is subject to a kind of levelling-down worry. In elucidating the equal opportunity rationale for exit initiatives, I suggested that senior cohorts are bigger beneficiaries of both vertical and horizontal opportunity-gaps – meaning that baby-boomers in advantageous socioeconomic positions are likely to enjoy much more diachronic opportunity with regards to their disadvantaged counterparts and the cohorts which have followed them. However, by aiming to reduce these horizontal and vertical opportunity-gaps with exit initiatives, it might be alleged that we are simply levelling-down aspirations and opportunities for advantageous positions, rather than raising standards and incentivising the creation of more opportunities for younger cohorts.

In response, it should first be noted that entry initiatives work by a similar mechanism insofar as they redistribute opportunities for advantageous positions without necessarily lowering aspirations or opportunities in the aggregate. In both cases, ex-post initiatives seek to correct unjust disparities with regards to equality of opportunity by positively discriminating during, or following a competitive procedure. Secondly, and perhaps most crucially, my argument does not rest on a telic commitment to eliminating all inequality at any cost, but rather on a deontic concern with ensuring that inequalities of outcome (with regards to income, wealth, respect, and fulfilment) do not stem from inequalities of opportunity. ³⁶ The forward-looking rationale for exit initiatives is not to reduce the number of opportunities overall in the name of promoting equality, but to ensure that access to them is structured in a way that fosters more dynamic and open competition across generations. The aim is to counteract barriers to mobility that have been reinforced by the compounding advantages accrued by senior cohorts, rather than to diminish opportunity in absolute terms.

Advantaged senior role-occupants have historically benefitted from structural injustices, such as racism, sexism, or other entrenched inequalities, and this has produced bigger horizontal opportunity-gaps within their cohorts. Moreover, successive governmental policies have failed to account for shifting macroeconomic conditions, leading to greater social immobility and the widening of vertical opportunity-gaps between cohorts. As a result, exit initiatives do not serve to level-down, but function as a means of recalibrating the distribution of opportunities over time, ensuring that younger cohorts are not unduly constrained by the cumulative advantages enjoyed by their predecessors. Instead of reducing opportunities, they aim to restore the conditions for intergenerational fairness and greater diachronic equality of opportunity. ³⁷

In short, the presumption in favour of entry initiatives requires further defence and there appear to be no immediately decisive objections to the institutionalisation of exit initiatives. If one endorses the equal opportunity rationale for affirmative action, then this will generate moral reasons in favour of exit initiatives in at least some cases and thus shift or split the burden of affirmative action as a consequence. Additionally, whilst the kind of exit initiatives discussed here constitute one way of shifting this burden, this paper should also serve to highlight a justification for other policies which have previously received little attention. Even though the aforementioned presumption requires further defence, one might find the proposal of exit initiatives to be unpalatable for one reason or another. This, however, still leaves us with a number of potentially more palatable redistributive policies open which merit further consideration. The same rationale, for instance, could be used to justify less abrasive exit schemes which involve senior role-occupants sharing income or dividends towards the end of their careers in order to create additional junior positions which could then be subject to complementary entry initiatives. ³⁸ In any case, I hope to have elucidated the equal opportunity rationale for exit initiatives and successfully challenged the status quo presumption in favour of entry initiatives by rebutting a number of potentially decisive objections.

Conclusion

In this paper, I have sought to challenge the status quo presumption against exit-based affirmative action by elucidating how, in at least some cases, we have as much or greater reason from equality of opportunity to enact what I refer to as mixed exit initiatives. I then considered a series of potentially decisive objections to exit initiatives as I have construed them and argued that these objections largely bias the status quo or constitute reasons to reject ex-post initiatives altogether when given considerable weight. Supposing that some readers still find the enforcement of exit initiatives to be unpalatable in spite of this rebuttal, I have also gestured towards alternative schemes to share income or dividends which may achieve similar results and merit further consideration but which stop short of lay-offs. For those who remain unconvinced of my conclusion here but who, for all intents and purposes, endorse the equal opportunity rationale for affirmative action, I realise that my argument may constitute an inadvertent attempt at a kind of reductio ad absurdum. Although this was not my intention, readers who arrive at this conclusion should perhaps reflect upon the reasons they have for endorsing affirmative action with a view to refining their justifications or objections further, and I hope that this paper at least proves useful to them in this respect.

That said, I urge advocates of the equal opportunity rationale for affirmative action to reexamine any presumptions they might share against exit initiatives. My central claim is not that entry initiatives ought to be abandoned or that they do not promote equality, but that focusing on them exclusively ignores the significant inequalities of opportunity which exist across different cohorts and within senior cohorts. By implementing mixed exit initiatives of the kind discussed, institutions may be better placed to address both horizontal and vertical opportunity-gaps. Such an approach not only promotes equality of opportunity within senior cohorts or between advantaged and disadvantaged groups across different cohorts, but also has the upshot of more equitably distributing the burdens of affirmative action across the broadest shoulders. As we continue to confront the disparities which the relevant schemes are designed to redress, it is important that we consider the disadvantages which cohorts face as socially salient groups and how these disadvantages compound at their intersection with other protected characteristics.