I. Introduction
Since the economic crisis of 2008, think tanks have pursued research into the causes and effects of socioeconomic mobility;Footnote 1 columnists and bloggers like Paul Krugman, Tyler Cowen, and Reihan Salam have debated whether mobility is desirable;Footnote 2 and politicians — from both ends of the political spectrum and both sides of the Atlantic — have emphasized mobility’s importance.Footnote 3 In contrast, socioeconomic mobility has received little attention from philosophers, who have instead focused their attention on social and economic inequality.Footnote 4
This essay will examine the connection between socioeconomic mobility and equality, and argue for two conclusions:
(a) Socioeconomic mobility is conceptually distinct from three common species of equality: (1) equality of opportunity, (2) equality of outcome, and (3) relational equality.
(b) However, socioeconomic mobility is connected — in different ways — to each species of equality, and, if we value one or more of these species of equality, these connections endow it with derivative normative significance.Footnote 5
I bypass two other, closely related issues in this essay, for reasons of theme and length, but flag them here to avoid confusion. The first is whether socioeconomic mobility has normative significance for reasons unrelated to equality: either in its own right, or via its connection to values other than equality. I take no position on that question here, though I explore the relation between mobility and other values, such as hope and democracy, in other work. The second is whether and why we should value any of the species of equality I discuss — equality of opportunity, equality of outcome, or relational equality. This is well-trodden ground. I do not aim in this short essay to persuade egalitarians to value one of these forms of equality over others, nor to persuade nonegalitarians to value equality at all. My goal here is more modest: to show that, if you value any of these species of equality, you should care about mobility.
Section II differentiates several forms of mobility, and briefly reviews relevant findings from the social sciences. Section III investigates the relationship between mobility and equality of opportunity. Section IV focuses on mobility's relation to equality of outcome. Section V examines the connection between mobility and relational equality.
II. Defining and Measuring Mobility
A. Defining mobility
All forms of socioeconomic mobility involve changes in an agent’s stock of some good — such as income, wealth, or social standing — over time. Footnote 6 Measuring socioeconomic mobility requires tracking these changes with respect to a reference point: upward mobility involves an increase in the individual’s stock of goods with respect to the reference point, whereas downward mobility involves a decrease. Choices about whose goods to track, which goods to track, what time period to consider, and what point of comparison to use differentiate several forms of mobility:
• Choices about whose goods to track differentiate intragenerational mobility, which involves changes in individuals’ circumstances, from intergenerational mobility, which involves changes in a family’s circumstances.Footnote 7 They also differentiate group mobility from individual mobility.
• Choices about which goods to track differentiates economic mobility, which involves changes in wealth or income; social mobility, which involves changes in occupational status, political power, or public esteem; and other forms of mobility such as educational or housing mobility.Footnote 8 While different forms of mobility are often closely correlated — upward occupational mobility, for instance, often produces upward wealth mobility — they involve the measurement of different goods.
• Choices about the time period to use also differentiate intragenerational from intergenerational mobility. In addition, they differentiate long-term mobility from short-term mobility.
• Perhaps most important from a normative perspective, choices about the reference point differentiate absolute mobility, which uses individuals’ own past holdings as a reference point, from relative mobility, which uses others’ holdings as the reference point.Footnote 9 Upward relative mobility for some implies downward relative mobility for others: if one person moves from a lower decile to the top decile of incomes, someone else in the top decile must drop into a lower decile.
Because the types of equality discussed in Sections III and IV involve individuals' comparative access to opportunities or outcomes, my focus will initially be on relative rather than absolute mobility. Absolute mobility contributes most clearly to relational equality, which I discuss in Section V.
B. Measuring mobility
Mobility is harder to measure than equality. Obtaining data about individual mobility requires a panel study, which surveys the same individuals at different times and therefore presents the challenge of reidentifying and reenrolling participants.Footnote 10 However, data on mobility has been collected despite these challenges. I will not provide an extensive survey of empirical research on relative economic mobility here. Interested readers should consult a comprehensive review.Footnote 11 Here, however, are some facts about mobility in the United States:
• Intergenerational and intragenerational relative income mobility are limited. Where a person ends up in the social and economic hierarchy depends heavily on her starting place and that of her family.Footnote 12 The same is true for relative wealth mobility.Footnote 13
• Intergenerational relative income and wealth mobility upward from the lowest quintile and downward from the highest is lower than mobility in and out of other quintiles. Forty-two percent of children starting in the lowest income quintile remain in that quintile as adults, while only 9 percent starting at the top drop to the bottom; meanwhile, 39 percent of children starting at the top stay there, while only 6 percent rise from the bottom to the top.Footnote 14 This shows that population-level mobility rates can obscure important differences in mobility within the population, just as — for instance — measuring the average flow across a wide river may obscure differences between channels of fast-moving water, slower-moving areas, and eddies where the direction of flow may become turbulent or even reverse direction.Footnote 15
• Intergenerational relative income mobility in the United States has not increased in the past thirty years, and some believe it has decreased.Footnote 16 This suggests that (if we think relative mobility is too low) we are not in a temporary downturn but rather in a situation that requires substantial change.
• Intergenerational and intragenerational relative income mobility differs substantially across countries.Footnote 17 This indicates that current levels of relative mobility in the United States do not reflect some more fundamental limit on mobility.
• The majority of families experience intragenerational absolute upward income mobility: they make more money at midlife than they did when they entered the workforce.Footnote 18 Most families also experience intergenerational absolute upward income mobility: family heads make more money than their parents made.Footnote 19 But rates of absolute mobility are slowing.Footnote 20
• However, the magnitude of this absolute mobility is fairly small: median yearly family income increased by only $3,772 (17 percent) from 1974 to 2004.Footnote 21 And this mobility is due to women working longer hours, rather than workers earning more per hour: median hourly compensation across all workers rose only 10.7 percent from 1973 to 2011 even though average hourly compensation grew nearly four times as much,Footnote 22 and men thirty to thirty-nine years old had lower median income in 2004 than in 1974.Footnote 23 So, although families have enjoyed modest upward absolute mobility, well-off workers have enjoyed more than others. Median income is also not keeping pace with productivity, as it did from 1945 to the 1970s.Footnote 24
II. Socioeconomic Mobility and Equality of Opportunity
Philosophers, social scientists, and policymakers have often equated mobility with equality of opportunity, which may explain philosophers’ comparative inattention to mobility.Footnote 25 In this section, I will argue that relative mobility serves as a valuable signal that equality of opportunity exists, but that the two concepts are distinct.
I will employ Rawls's definition of “equality of opportunity.” Under Rawlsian equality of opportunity,
we might say that those with similar abilities and skills should have similar life chances. More specifically, assuming that there is a distribution of natural assets, those who are at the same level of talent and ability, and have the same willingness to use them, should have the same prospects of success regardless of their initial place in the social system. In all sectors of society there should be roughly equal prospects of culture and achievement for everyone similarly motivated and endowed. The expectations of those with the same abilities and aspirations should not be affected by their social class.Footnote 26
Winship (in this volume) defines equality of opportunity similarly: “inequality of opportunity exists when some are more able than others to achieve their preferred life.”Footnote 27 For both Winship and Rawls, equality of opportunity entails more than the absence of discrimination or legal barriers to opportunity, but it does not entail unlimited freedom of opportunity: a society can realize equal opportunity even if people have no opportunity to be bank robbers. Rather, equality of opportunity implies equal access to the social and economic positions that exist in society.
It is easy to see how relative mobility might correlate with equality of opportunity: for instance, if children who start in different social classes reach the top quintile at similar rates, this suggests that social class does not affect children's opportunities to reach the top quintile. In contrast, while — as Winship observes — absolute upward mobility may improve opportunity levels for disadvantaged children,Footnote 28 absolute mobility does not indicate that their opportunities are more equal. And shifting the focus away from equality of opportunity toward opportunity levels is not an interpretation of equality of opportunity, but an alternative to that ideal.Footnote 29
If we normatively value the kind of equality of opportunity that Rawls and Winship discuss, we should therefore care about the existence of mobility in society. (Recall that I do not attempt here to provide an account of why we should value this kind of equality of opportunity. Some may treat equality of opportunity as valuable in itself, while others may treat it a signal that some deeper underlying value is being fulfilled.Footnote 30) However, mobility is not equivalent to equality of opportunity, nor is mobility a constituent of equality of opportunity. Rather, mobility is a signal that equality of opportunity is present: it correlates loosely with equality of opportunity but does not constitute it.
First, mobility is not a necessary consequence of equal opportunity, because observed mobility levels depend not only on what opportunities individuals have available but also on which opportunities they prefer to take. Low levels of mobility could be attributable to differing preferences rather than unequal opportunities: if, say, people in a cosmopolitan community pursue opportunities for career advancement while those in a conservative community prefer not to.Footnote 31 This example indicates that mobility is not a component of equal opportunity: rather, low mobility is a defeasible signal that opportunities were unequal. Immobility-producing preferences are one potential defeater to the inference from low mobility to unequal opportunity.
Second, no specific level of mobility — not even “perfect mobility,” which exists where a person’s past status or parents’ status does not correlate at all with their present statusFootnote 32 — is sufficient, without more, to show that equal opportunity exists. For example, perfect mobility in a society where middle-class children, on average, are much more talented and able than the children of rich parents would not demonstrate the existence of equal opportunity, since equal opportunity in such a context might in fact produce a negative correlation between parents’ and children’s status. The overarching point is that, since mobility data provide us information about correlations in status but not about the specific processes that produce those correlations, observed mobility levels (whether low or high) cannot by themselves demonstrate the presence or absence of equal opportunity.
Third, equality of opportunity (in either Rawls’s or Winship’s sense) depends both on what opportunities people have to move between disparate positions and how disparate those positions are. Immobility is a signal that individuals lack opportunities to move between positions, but if the available positions are very similar, opportunity can be highly equal even though opportunities to move are nearly absent. To take the most extreme case, if there is no labor specialization and therefore only one social position in a given society, there is equal opportunity even though there is no mobility.Footnote 33 This outcome, as Winship observes, may produce a low level of opportunity, but is consistent with equality of opportunity.Footnote 34
Finally, whether intragenerational or intergenerational relative mobility is relevant to equality of opportunity depends on how we understand equality of opportunity. If we interpret equality of opportunity as requiring only a “moment of equal opportunity,”Footnote 35 as Benjamin Sachs suggests we should, then equality of opportunity does not require intragenerational mobility; there is no reason to worry about a society where very few workers move up or down over a given ten-year period, as long as they all at some point — for example, at age eighteen — enjoyed an equal opportunity to compete for advantaged positions.Footnote 36 However, Sachs's interpretation of equality of opportunity connects equality of opportunity with intergenerational mobility: if someone’s income in childhood strongly correlates with his or her income in adulthood, this raises (defeasible) concerns about whether the children of the poor and the children of the rich ever enjoyed equal chances. In contrast, if we interpret equality of opportunity as applying “at each and every stage in a person’s life,”Footnote 37 then some degree of intragenerational mobility will correlate with equality of opportunity.
Ultimately, relative mobility is normatively important as an indicator of certain forms of equality of opportunity. Although we can have mobility without equality of opportunity, and equality of opportunity without much mobility, mobility represents a useful and observable proxy for equality of opportunity. But relative mobility is not equivalent to equality of opportunity, nor does it produce equality of opportunity.
III. Socioeconomic Mobility and Distributive Equality
Part II argued that relative mobility is a proxy for, rather than a cause of, equality of opportunity. The relationship between relative mobility and distributive equality is different: relative mobility is not just a proxy for distributive equality, but can also help ensure that goods are distributed more equally over the long term. Therefore, if we value distributive equality, we should care about relative mobility not as a proxy but instead as a way of realizing distributive equality.
As with equality of opportunity, a comprehensive defense of the normative importance of distributive equality is outside the scope of this project.Footnote 38 As with inequality of opportunity, some may regard distributive inequality as objectionable in itself, while others may believe that certain forms of distributive inequality hamper the achievement of other values. Milton Friedman seems to take the latter view when he asks us to:
[c]onsider two societies that have the same distribution of annual income. In one there is great mobility and change so that the position of particular families in the income hierarchy varies widely from year to year. In the other, there is great rigidity so that each family stays in the same position year after year. Clearly, in any meaningful sense, the second would be the more unequal society. The one kind of inequality is a sign of dynamic change, social mobility, equality of opportunity; the other, of a status society.Footnote 39
Although Friedman seems to prefer the first, mobile society to the second, rigid one, he may regard the inequality in the rigid society as objectionable, not in itself but because it is indicative of an undesirable “status society.” One reason a status society may be objectionable involves relational equality, which I discuss in Section V.
There are two ways of minimizing long-term distributive inequality:
1) Minimize distributive inequality over short time scales;
2) Permit distributive inequality over short time scales, but ensure — as in Friedman’s first, mobile society — that levels of relative mobility are high.
A society with unequal positions but frequent mobility between positions will exhibit high inequality at any given time — some people will be business leaders while others are janitors — but low inequality when viewed over a longer time scale.Footnote 40 The Economic Mobility Project offers a stylized example, similar to Friedman’s, to illustrate how relative mobility reduces long-term inequality:
Think of a hotel in which some of the rooms are luxurious executive suites while others are small and modest. The executive suites may be getting fancier over time and the modest rooms ever more modest. But if a different group of people occupies the executive suites each year, and everyone has a decent shot at staying in these fancier rooms, people have less reason to complain. Relative mobility is similar to this kind of room-changing. In particular, if relative mobility had increased at the same time that income inequality has risen, then there would be less reason for concern about rising inequality.Footnote 41
Because people move from janitorial to leadership positions, or because janitors’ children grow up to be leaders, incomes and social positions in a society with high levels of mobility will converge over long time scales, as when we compare individuals’ lifetimes (if we have high intragenerational relative mobility) or the history of family lines (if we have high intergenerational relative mobility).Footnote 42
But, recall that establishing and maintaining short-term distributive equality is also a way of realizing long-term distributive equality. Why should a society that cares about distributive equality invest in promoting relative mobility — increasing movement up and down the ladder — rather than focusing its efforts on bringing the rungs closer together by promoting short-term equality? The most persuasive case I can imagine for pursuing distributive equality via relative mobility rather than via short-term equality rests on the efficiency of labor specialization and organizational hierarchy. If Ann specializes in baking bread and Beth specializes in cheesemaking, each is more efficient than she would be if she had to engage simultaneously in both forms of labor.Footnote 43 Similarly, social arrangements that allow for some hierarchy in civil society, employment, and governance are able to function more effectively than those lacking managers or leaders.Footnote 44 In contrast, short-term equality of status and income does not allow for specialization or hierarchy.
Relative mobility, then, represents a path toward a potential reconciliation of efficiency with distributive equality, because it permits us to achieve distributive equality without needing to ensure short-term equality. Paul Gomberg and Andrew Sayer have recently argued that fairness requires the abandonment of specialization and hierarchy,Footnote 45 which — if the empirical assumptions above are correct — would greatly diminish efficiency. Adopting their proposals would lead to our sharing work more equally, but our productivity dropping sharply. In contrast, relative mobility allows us to maintain high levels of productivity while widely distributing — at least over a long time horizon — the benefits of this productivity. In a society with relative mobility, even if cheesemakers currently earn more than bakers, bakers’ children are not locked into becoming bakers themselves, nor are cheesemakers guaranteed their current high incomes or bakers locked into their career for a lifetime. Similarly, those who currently occupy high positions in an organizational hierarchy do not enjoy their positions for an entire lifetime, nor can they pass their status on to their children. Relative mobility will in time move others up into these positions, where they, too, can for a time enjoy the higher status that accompanies them.
Additionally, short-term equality may be inefficient given the indivisible nature of existing goods in society. For instance, strict short-term equality in housing or education implies that no one may live in elegant houses that not all may occupy, or attend exceptional universities not simultaneously accessible to all. In contrast, relative mobility can allow us to benefit from these indivisible goods while dividing up their benefits more equally than a society with fixed hierarchies would.Footnote 46
The advantages to the distributive egalitarian of pursuing distributive equality by way of relative mobility rather than moment-by-moment equality are now clear: relative mobility enables society to achieve distributive equality, at least in the long term, without having to level down indivisible goods or give up on the benefits of specialization. The question is whether achieving distributive equality via relative mobility has any disadvantages when compared to achieving it via short-term equality. I will discuss two such potential downsides. The first involves loss aversion: relative mobility may be psychologically difficult to accept. The second involves the possibility that short-term equality matters in itself.
A. Should we cater to loss-averse preferences?
Social scientists discussing the ability of relative mobility to mitigate inequality note that a society with frequent mobility can be psychologically burdensome, particularly on those who fall from advantaged positions.Footnote 47 These observations often appeal to the psychological phenomenon of loss aversion to argue that relative mobility is undesirable from a welfarist standpoint.Footnote 48 We therefore face a trilemma: either we must abandon distributive equality, accept the inefficiencies associated with short-term equality, or endure the psychological stress that relative mobility causes.
One approach to this trilemma — one which involves adopting a welfarist point of view — is empirical: social scientists could attempt to measure whether mobility’s fulfillment of inequality-averse preferences and its contribution to efficiency produces enough psychological satisfaction to compensate for its frustration of loss-averse preferences. Here, it is worth noting that there may be psychological gains from mobility — if people experience psychological satiation after remaining at the same level for some time, then a life that involves upward and downward mobility may actually be welfare enhancing.Footnote 49 This empirical approach faces a number of problems. Measuring psychological satisfaction is empirically difficult. And it is not obvious that maximizing psychological satisfaction on a population scale is a normatively desirable goal: the well-off have the most to lose from relative mobility, whereas the worst-off stand to gain.Footnote 50
I will take a different approach, which queries the normative importance of loss aversion rather than focusing on its prevalence or psychological effects. There is broad agreement that satisfying certain individual preferences is not an appropriate goal of public policy. These include unjust preferences (e.g., the pleasure a racist receives when minorities are subject to invidious discrimination), as well as — though more debatably — external preferences (preferences about how someone else’s life goes) and tuistic preferences (preferences that involve comparing one’s own well-being or stock of goods to others’).Footnote 51 Loss aversion seems structurally similar, though not identical, to tuistic preferences such as envy. Both John Rawls and David Gauthier have argued that social policy should not cater to envious preferences: rational individuals care about their own holdings, not the extent to which their holdings exceed others'.Footnote 52 Similarly, it might seem, rational individuals care about their own holdings of goods over a lifetime, not the extent to which their holdings exceed their own past holdings, their parents’ holdings, or others’ holdings. As such, the satisfaction of loss-averse preferences, like the satisfaction of envious preferences, would seem an inappropriate object of public policy. This is so particularly if we imagine the task of public policy — as Rawls does — as providing all-purpose goods to individuals in society rather than catering to their psychological well-being.Footnote 53 Although one might worry that the case for distributive equality itself rests on tuistic preferences, there are reasons to care about distributive equality that do not appeal to tuistic preferences: distributive equality may be a good in itself, may be valuable in virtue of its contribution to welfare given the diminishing marginal utility of resources, or may be a proxy for some other value such as relational equality.
To the extent that we remain unpersuaded by the above normative argument against counting loss-averse preferences,Footnote 54 we face the problem of balancing the burden of loss aversion against the benefits of specialization. Absolute upward mobility can be relevant here, as Benjamin Friedman has argued: according to Friedman, even if relative mobility drops individuals into a lower income decile, they will not experience that relative drop as psychologically burdensome as long as they also move upward in absolute terms.Footnote 55 This would then provide an additional reason to value upward absolute mobility: not only does it improve well-being, but it also makes relative mobility possible. Another way of minimizing the effects of loss aversion might involve stressing the fact that individuals’ social and economic positions are subject to change, in order to discourage psychological attachment to one’s present position.Footnote 56 A final approach would be to couple relative mobility with a generous social minimum, so that downward mobility will not provoke anxiety about destitution.
An additional reason to doubt the normative relevance of catering to loss-averse preferences rests on a concern about loss aversion being collectively self-defeating. Even if each person is loss-averse and so individually prefers a lower level of relative mobility, each person (or, at least, society as a whole) may be better off, even counting the psychological burdens of loss aversion, if the level of relative mobility in society is higher than they would have chosen for themselves. Such an argument would parallel Robert H. Frank’s case for collective regulation of positional goods: even if each person would like to be positionally better off than others, we may prohibit certain forms of positional competition because they leave us all worse off.Footnote 57
B. Does short-term distributive inequality matter?
Dennis McKerlie asserts that short-term equality is valuable in itself:
[I]magine a new kind of egalitarian society. It contains great inequality, with happier lives attached to certain social positions. But at a fixed time people change places and switch from a superior position to an inferior one or vice versa. One example would be a feudal society in which peasants and nobles exchange roles every ten years. The result is that people's lives as wholes are equally happy. Nevertheless during a given time period the society contains great inequality, and in one sense this always remains true. I will call this system ‘changing places egalitarianism.’ If equality between complete lives were all that mattered, an egalitarian could not object to it. But I think that many egalitarians would find it objectionable.Footnote 58
Although McKerlie does not use the term “relative mobility,” his view would support the importance of short-term equality, and would therefore show that distributive egalitarians have reason to select short-term equality as the way of realizing long-term distributive equality (“equality between complete lives”).
The feudal society may be objectionable because it violates relational equality, as I will discuss in Section V, but it is very hard to see why short-term distributive inequality between people matters. McKerlie himself admits as much:
If two people will see a dentist tomorrow, [concern about short-term equality] would tell them to schedule simultaneous appointments so that there will be equality in suffering at that time. Are there serious egalitarian reasons for preferring two 10:30 appointments to an appointment at 10 and an appointment at 11?Footnote 59
McKerlie concedes that there is no reason to choose simultaneous appointments, but does so for the wrong reason, concluding that inequality lacks moral significance if “the time periods within which the inequality is measured are too short.”Footnote 60 But lengthening the time period doesn’t make short-term inequality any more objectionable. Even if rehabilitating a broken leg takes a year, two people breaking their legs in different years seems no more objectionable than their breaking their legs in the same year. Although short-term insufficiency may count against the pursuit of efficiency or distributive inequality,Footnote 61 this reason appeals to humanitarian considerations rather than distributive ones.
Does the broken leg example prove too much — does it show that long-term distributive inequality is patently irrelevant when it does not involve relational inequality? I do not believe so. It seems plausible — though not decisive — that one person suffering two broken legs is worse than two people each suffering one.Footnote 62 If we have to decide whom to save from a broken leg, it makes sense to save the person who has already had to suffer a broken leg. (The pull of distributive equality, however, is weak in comparison to that of well-being: if we can only save one of two people from a broken leg, we should do it. That both of their legs being broken would better realize distributive equality is irrelevant.)
Ultimately, then, relative mobility should matter to distributive egalitarians. It enables the realization of long-term distributive equality while allowing for the efficiency of labor specialization and indivisible goods. While relative mobility is not the only way of achieving long-term distributive equality, it is the best way of achieving long-term distributive equality without sacrificing well-being. Relative mobility can contribute to distributive equality.
IV. Socioeconomic Mobility and Relational Equality
McKerlie's example of nobles and peasants raises a third question — one that he overlooks — about mobility: What implications does relative mobility have for relational equality, which (in Elizabeth Anderson's version) focuses on providing each citizen the capabilities needed to avoid oppressive social relationships and to participate as an equal citizen in a democracy?Footnote 63 In this section, I will consider two questions: (a) whether short-term inequality plus relative mobility is compatible with relational equality, and (b) whether absolute mobility is sufficient for relational equality.
A. Is short-term inequality plus relative mobility compatible with relational equality?
Some of McKerlie's critics have argued that our response to the nobles and peasants case does not reflect worries about short-term inequality, but rather concerns about relational inequality.Footnote 64 Another example of McKerlie’s also raises the question of relational equality:
Most of us believe that there should be equality in a marriage. A marriage in which the husband makes all of the important decisions, and achieves his own goals with his wife's help, can be criticized for not being an equal partnership. Rather than having the inequality continue throughout the marriage, it would be better if the wife had her turn at being the leading partner — if, later in the marriage, she were able to achieve her own goals with the assistance of her husband. Then, if we think about the marriage as a whole, the husband and wife have equal shares in decision-making and responsibility. Even if the marriage does change in this way, we would not regard it as the best kind of marriage. It is not an ideal marriage because the husband and wife are never equal partners during the marriage. There is no time at which they share decision-making and responsibility. The fact that their shares of these goods accumulated through time are equal is important, but it does not make up for this deficiency.Footnote 65
McKerlie draws the wrong moral in the marriage case, concluding that the marriage example shows “that matching a present inequality with an inequality at some other time in favor of the other person does not destroy the significance of the current inequality.”Footnote 66 But the oppressive nature of the relationship between husband and wife, or noble and peasant, may worry us even if — as I have argued — the short-term inequality should not.
The question, then, is whether a society with short-term inequality but high relative mobility will entail the objectionable relational inequality that exists between nobles and peasants or between McKerlie’s imagined spouses. The answer, I believe, depends on the details of the society’s economic structure. Imagine three different structures:
Mobile Entrepreneurs: Workers’ income and status vary from year to year, but workers are all entrepreneurs in small businesses with flat hierarchies — higher-income workers do not exercise managerial authority over others.
Mobile Bureaucracy: Workers take turns functioning as managers and as managed employees. When workers are in managerial roles, their income and status is higher; when workers are in nonmanagerial roles, their income and status is lower.
Crony Capitalism: Workers are divided into groups whose income and status varies dramatically from year to year depending on which group controls political and economic power. No group invariably dominates the struggle, but the group in charge at any given time exercises domineering authority over the others.
I see no immediate reason to find Mobile Entrepreneurs objectionable: workers enjoy long-term distributive equality as well as equality of social standing and power. In contrast, Crony Capitalism — like McKerlie’s example of peasants and nobles — seems objectionable. The most complicated case is Mobile Bureaucracy. Here, the workers moving up and down stand in power relations over each other, unlike in Mobile Entrepreneurs, but these relations are not obviously oppressive, as they are in Crony Capitalism. I am inclined to believe that managerial power — which arguably is a major source of corporate efficiency — does not inevitably constitute oppression, though this depends on the details of workplace governance.Footnote 67 As long as managers are accountable to their employees, the manager/employee distinction should be compatible with nonoppressive social relations.
However, even short-term dispersion of income, wealth, or status between different individuals in society — if it becomes too extreme — might itself undermine relational equality and thus render cases like Mobile Entrepreneurs objectionable. Consider, for instance, the concern Joshua Cohen raises about “a community with no very needy members” in which, nonetheless,
[p]aying the incentives that are necessary, given de facto preferences, to maximize the advantage of the least advantaged results in extreme income dispersion: a dispersion that we are inclined to find objectionable or excessive, either because of the preferences that produce it or, more likely, just because it strikes us as objectionable that some people live that much better than others.Footnote 68
Cohen asserts that “[l]arge dispersion cases [are] objectionable in part because the sheer magnitude of differences may be taken as indicating lack of respect.”Footnote 69
Why, exactly, does Cohen think extreme dispersions are objectionable? Their objectionable nature does not stem from the absolute disadvantage of the worst-off members, since no one is very needy. (And, therefore adding absolute upward mobility will not solve the problem.) Nor is the objection grounded in distributive inequality: a smaller income dispersion would be unequal but would not be objectionable. Nor is it grounded in the tuistic preferences of the worst-off: Cohen relies on Rawls — who rejects tuistic preferences — for the claim that some large dispersions “cannot help but cause a loss of self-esteem.”Footnote 70 Even when large dispersions produce envy, Rawls believes such envy is different from ordinary envy: it is the result of the direct loss of self-respect, rather than the cause of that loss.Footnote 71 So, Cohen's objection — as I reconstruct it — is that extreme income dispersions are not only evidential of relational inequality but also directly expressive of relational inequality.
Another relational egalitarian, Debra Satz, agrees that relational equality has comparative elements.Footnote 72 In her view, the capacity to function as an equal citizen depends not only on one’s own knowledge, but also on how one’s knowledge compares to that of others.Footnote 73 Extreme disparities in knowledge are therefore inconsistent with relational equality.
It remains unclear how far Cohen’s and Satz’s concern applies to extreme short-term rather than extreme long-term income dispersion. In Mobile Entrepreneurs, although a large gap exists at any given moment between the successful and the unsuccessful, other social mechanisms are working effectively to ensure that people move up and down intergenerationally and intragenerationally. And everyone in Mobile Entrepreneurs enjoys a roughly equal lifetime income that is higher they would have enjoyed in a society with greater short-term equality. Under these circumstances, perhaps extreme moment-by-moment income dispersions will be less expressive of disrespect: individuals doing comparatively less well now can be assured that they, or their children, will do better soon, and those momentarily doing very well cannot regard themselves as forming an enduring class that is superior to others. There will not be knowledge that some people or families are forever unable to access. Yet people will still stand in starkly different positions as they interact in day-to-day life, and this in itself may be destructive of relational equality.
Evaluating the normative status of societies like Mobile Entrepreneurs is practically pressing given actual economic trends away from a “diamond” pattern of income distributions with a large number of individuals near the median income toward an “hourglass” pattern with many fewer middle-class individuals.Footnote 74 Does relational equality require preserving a “diamond” structure with many middle-class jobs?Footnote 75 Or can relational equality be achieved even in an “hourglass” society so long as we have high relative mobility of individuals into and out of the top positions of the hourglass?
My ultimate answer to this last question — and to the initial question of this subsection — is equivocal. If we find Cohen’s concern about large dispersions compelling, and believe it extends to large short-term dispersions, then we should deny that an “hourglass” society can achieve relational equality by way of relative mobility, and instead believe that relational equality also requires a range constraint on short-term inequality that prevents extreme dispersions from arising.Footnote 76 But if we either don’t find that concern compelling or believe it only applies to extreme long-term dispersions, relational equality should be achievable, via relative mobility, in cases like Mobile Entrepreneurs , and in “hourglass” societies more generally. Note that relative mobility would still not be sufficient on its own for relational equality: relational equality also requires civic and political equality, topics which go beyond my scope here.
B. Can we have relational equality without relative mobility?
Reihan Salam and Alex Tabarrok have argued that society should aim to increase upward absolute mobility and should treat relative mobility as unimportant.Footnote 77 While absolute mobility can contribute to relational equality, relational equality requires relative mobility. We can only give up on relative mobility if we are willing to give up on relational equality altogether.
As Tabarrok notes, upward absolute mobility is compatible with a complete lack of relative mobility.Footnote 78 But, as Friedman’s observations suggest, it is hard to see how a rigidly stratified society without relative mobility could be compatible with relational equality. Upward absolute mobility can also be unequally distributed through the population, and can coexist with oppression and political inequality: society could enjoy upward absolute mobility even though its best-off members comprise an entrenched ruling class while those worse off advance only marginally.Footnote 79 Finally, many types of absolute mobility — such as the substantial improvements in access to consumer appliances that Steven Horwitz emphasizes in this volume — contribute only modestly to relational equality.Footnote 80 Absolute upward mobility could enable both Rich and Poor to own refrigerators without protecting Poor against rising child care expenses that block her career advancement,Footnote 81 intrusive governance at work,Footnote 82 and legislative enactments that cater only to Rich's interests,Footnote 83 all factors which preclude relational equality. (Although Horwitz notes that housing, health care, and educational expenses have risen, he does not discuss rises in child care or transportation expenses, nor does he discuss the shift from defined-benefit to defined-contribution pensions.Footnote 84)
The inability of upward absolute mobility — particularly upward absolute consumption mobility, on which Horwitz focuses — to secure relational equality on its own does not render it irrelevant. Poor people having refrigerators is a good thing: relational equality is not all that matters. But it does show that we need more than upward absolute mobility in order to achieve relational equality. We need civic and political equality, and we also need relative mobility. Anderson's version of relational equality, for example, explicitly requires mobility in and out of elite positions in society — that is, relative mobility.Footnote 85 The same is true for Satz’s account.Footnote 86
Short-term egalitarians might raise a final question. Even if relative mobility is required for a society with elite positions, could a society with strict short-term distributive equality and no elite positions — like those Gomberg and Sayer advocate — achieve relational equality without relative mobility? I do not think so. Relational — as opposed to distributive —equality seems to require democracy and civil society, which entail specialization and hierarchy and therefore require relative mobility.
V. Conclusion
I have argued that, if we care about equality, we should care about mobility. And I have illustrated how different forms of mobility connect with different types of equality:
(a) Relative mobility is a proxy for equality of opportunity. Although relative mobility is neither necessary nor sufficient for equality of opportunity, the two are correlated.
(b) Relative mobility's relation to distributive equality is different: rather than being a proxy for distributive equality, it is a contributor to distributive equality, because it can convert short-term equality into long-term distributive equality.
(c) Relative mobility and absolute upward mobility are both contributors to relational equality, although absolute upward mobility is not sufficient on its own to achieve relational equality.
How might we promote the types of mobility that correlate with, or contribute to, equality? Some efforts — such as socioeconomic affirmative action policies — might serve to promote intergenerational relative mobility into and out of elite positions.Footnote 87 Others, such as policies incentivizing those currently poor to accumulate assets, would focus on mobility out of disadvantaged positions.Footnote 88
While much more work remains to be done in determining what values are at stake as we structure institutions, civil society, and informal individual life in ways that affect the frequency and magnitude of different types of mobility, I hope to at least have shown here that mobility's connection to equality raises questions of normative interest for moral and political philosophers. Rather than focusing solely on one-time measures of resources or status — whether, for instance, people possess equal resources in a given year, or over a lifetime — we might do well to also attend to the normative relevance of intertemporal changes in the distribution of resources and status.
