Summary:
· Equal opportunity is a post-Rawlsian approach to distributive justice. It considers that there are legitimate and illegitimate inequalities. According to economist and philosopher John E. Roemer, inequality may be legitimate if it results from individuals’ « efforts » (something within an individual’s control) but is illegitimate if it results from « circumstances » (something over which an individual has no control).
· According to Roemer, what constitutes « effort » or « circumstances » must be determined by society. However, it must be taken into account that the « effort » made by an individual is often largely determined by their « circumstances. » For example, a child’s attendance at school (i.e., their effort) depends on their social background. However, this variable is not under the control of an individual; it is a « circumstance. »
· In terms of income, equal opportunities are not achieved in France. In particular, there is a strong hierarchy in income distribution based on social background.
· In terms of health, less than 10% of inequalities are due to individual effort, even if we refuse to consider an individual’s « efforts » as being partially determined by « circumstances. » Approximately 40% are due to illegitimate « circumstances » (in particular, an individual’s family background).
There is a long tradition of dialogue between economists and philosophers who theorize about distributive justice (such as Rawls, Dworkin, Cohen, etc.). This article does not aim to present the entirety of these exchanges, but rather to illustrate some of the contributions made by economists, particularly on the subject of equal opportunity (or equality of opportunity). After introducing the concept of « equality of opportunity » as analyzed by economists (notably John E. Roemer (1998) and Roemer and Trannoy (2015)), the article will present some recent work by Alain Trannoy and his co-authors in order to show how empirical research can contribute to a reflection on distributive justice[2].
Equal opportunity
Equal opportunity is part of the post-Rawlsian philosophical thinking on distributive justice[3], which emphasizes individual responsibility.
In short, in this approach, inequalities between individuals must be analyzed according to their origins. Those due to individuals’ « efforts« can be considered legitimate and do not necessarily call for correction by society. On the other hand, those due to « circumstances, » i.e., factors beyond the control of individuals, are illegitimate. A « luck » factor can also be included in these analyses, although economic literature focuses primarily on « efforts » and « circumstances. »
Let us take the example of health inequalities—based on Jusot, Tubeuf, and Trannoy (2013) (the article will be presented in more detail later)—and imagine that two people have different medical statuses: one is in good health, the other is not. Is this inequality morally acceptable? And to what extent should society intervene to correct this inequality?
To answer this question, we can look for the origin of this inequality: is it due to circumstances beyond the individual’s control (a genetic factor, for example), in which case it would not be acceptable, or is it due to the individual’s efforts? For example, the « healthy » individual may exercise, refrain from smoking, eat a balanced diet, etc., while the « unhealthy » individual does not. In which case, the inequality would become more legitimate.
More rigorously, this literature puts forward a principle of compensation which establishes that two people who have made the same « effort » should have the same result (e.g., level of health, salary, etc.), regardless of their respective « circumstances. »
But how do we define « effort » and « circumstances »? John E. Roemer—one of the major contributors to this literature—believes that there can be no a priori definition of « effort » or « circumstances. » Each society must make its own choice. On the other hand, once « circumstances » have been determined, it is important to be consistent and take into account that they can influence the level of « effort » made by an individual. According to Roemer, only « efforts » that are not determined by « circumstances » constitute legitimate sources of inequality.
Roemer discusses, for example, the case of the « efforts » made by children at school. If these « efforts » result from family pressures, then if they later lead to inequalities (e.g., in wages), these inequalities are not fair: the effort ultimately stems from something for which the individual is not responsible (their family environment). According to Roemer, « effort » should be purged of the influence of « circumstances. »[7] Other views exist on this point, as we will explain below.
If we subscribe to this approach, public policies should seek to correct inequalities due to illegitimate sources – « circumstances » – so that they have no impact, or as little impact as possible, on individuals’ destinies. Ideally, they should also redistribute resources so that two individuals who have made similar efforts achieve the same results (in terms of health, education, income, etc.) regardless of their « circumstances, » i.e., ensure that the principle of compensation is respected. Of course, in practice it may be impossible to abolish differences due to family background, and the morality of such an objective is debatable (taken to the extreme, it would involve removing children from their families in order to eliminate family « circumstances »).
Formalized approaches to equal opportunity make it possible to devise tools for comparing « public policies, » ranking them, and determining the optimal policies for promoting equal opportunity. They also make it possible to clarify the different conceptions of this theory and to highlight the tensions between them. On this point, interested readers may refer to Marc Fleurbaey and Vito Peragine (2013—Trannoy 2016 also discusses this aspect in more detail). Rather than expounding on these points, the rest of this article presents, through two articles devoted to France, how empiricists have also appropriated these subjects.
Equal opportunities in France: income
A natural starting point for assessing equal opportunities is to consider individuals’ income and ask whether and to what extent « circumstances » affect a household’s income and, therefore, whether we can conclude that there is a breach of equal opportunities.
With this in mind, Lefranc, Pistolesi, and Trannoy (2009) study distributions of living standards [8] according to the social origin of the « head of the family » (were his parents farmers, employees, workers, etc.) using INSEE’s « Family Budget » surveys from 1979 to 2000.
They also discuss what « equal opportunities » means in terms of standard of living when three possible origins are attributed to this standard of living: « effort, » « circumstances, » and « pure luck »[9]. They also propose statistical tests to verify that the conditions necessary for equal opportunities are met.
Intuitively, in the case of equal opportunities, the social background of the head of the family should have no impact on the family’s standard of living, and these standards of living should be distributed identically. This provides a test for studying « equal opportunities » in relation to social background. However, this test may prove restrictive because, in order not to conclude that there is inequality of opportunity, all income distributions must be exactly identical (Lefranc, Pistolesi, and Trannoy (2009) analyze income distributions as a whole, not just an average value), which is very unlikely to happen in practice. The authors also examine « stronger » definitions of equal opportunity and seek to identify whether social background confers a « systematic » advantage in terms of income.
Using these tests, Lefranc, Pistolesi, and Trannoy refute the hypothesis of equal opportunity in France. On the contrary, they observe a strong social hierarchy, with some social origins strictly advantaged over others. While this hierarchy was not challenged during the period, the gaps between distributions narrowed slightly between 1979 and 2000. Lefranc, Pistolesi, and Trannoy therefore highlight (at least until the 2000s) a pronounced inequality of opportunity in France.
It should be noted that the World Bank is beginning to take an interest in this notion of equal opportunity. Roemer (2013) proposes a methodology for making international comparisons based on the idea of equal opportunity. In an illustration comparing individuals’ income distributions (their parents’ level of education is used to measure « circumstances »), France appears to be in the middle of the pack among 22 European countries. However, the data used by Roemer is (as he acknowledges) incomplete and sometimes questionable, which limits the scope of this result.
The results of Lefranc, Pistolesi, and Trannoy raise (at least) two questions: (a) Are these inequalities of opportunity only observable in terms of income? (b) Are the assumptions—implicit in this presentation but largely explained in the original article—about the consideration of effort, namely that effort must be purged of the impact of circumstances, really that decisive?
Equal opportunities in France: health
In an article published in 2013, Jusot, Tubeuf, and Trannoy develop an approach to measuring the share of health inequalities that stem from « circumstances » and « effort. » Another important point in this article is that it compares several approaches to distinguishing between legitimate effort and circumstances.
Indeed, Roemer’s approach developed above (and used in the previous income analysis) may seem extreme since it considers that « efforts » leading to legitimate inequalities are only those that do not in any way stem from « circumstances. » In terms of health, for example, an individual with healthy eating habits could not consider themselves « deserving » if these habits were instilled in them by their parents. Consequently, after measuring the level of effort made by an individual, an analysis based on Roemer’s view requires correcting this level of effort in order to purge it of the indirect effects of « circumstances. » For example, correcting the level of effort to remove the impact of the social environment on efforts.
Conversely, one could consider that an individual should be fully rewarded for their efforts, regardless of their origin. On this point, Jusot, Tubeuf, and Trannoy refer to Roemer’s book (1998), which recounts a controversy between Roemer and Barry. The latter defends the idea that even if an individual’s « efforts » (academic efforts in the controversy) are due to « circumstances » (e.g., family pressures), this does not make the individual less deserving of these efforts. In this case, we may wish not to correct the « efforts » of the « circumstances. »
Finally, Jusot, Tubeuf, and Trannoy present a final point of view, this time referring to Swift (2009). Let us imagine that we are observing both a parent and a child. How can we take into account the « efforts » of parents to improve the lot of their children? From the parent’s point of view, this is an effort, but from the child’s point of view, it is a circumstance (the child has no influence over these « efforts »). Swift gives ethical priority to family structures and considers that inequalities resulting from parental efforts are not unfair. This implies partially correcting individuals’ family « circumstances » for the « efforts » made by their parents.
By incorporating these alternative points of view, Jusot, Tubeuf, and Trannoy seek to compare the share of health inequalities linked to circumstances and efforts under three different modalities:
A. Roemer’s approach: « efforts » must be purged of « circumstances. »
B. Barry’s approach: « efforts » and « circumstances » are considered independent.
C. Swift’s approach: « circumstances » must be purged of the effect of « efforts. »
Using a health survey conducted in France in 2006, Jusot, Tubeuf, and Trannoy analyze the impact of « efforts » (defined as habits that impact health: smoking or not, eating vegetables, being overweight), « circumstances » related to the family environment (parents who smoke/drink, parents’ financial situations, etc.) and « demographic variables » (gender, age) on an individual’s perceived health. It should be noted that demographic variables are (by definition, as the authors acknowledge) « circumstances »: an individual has no influence over them and can hardly be held responsible for their gender or age. However, these « circumstances » are secondary in the analysis by Jusot, Tubeuf, and Trannoy, who seek primarily to measure whether approaches (A), (B), and (C)—which focus mainly on the influence of the family environment—lead to very different measures of inequality of opportunity.
Jusot, Tubeuf, and Trannoy find that between 44.5% (Swift’s approach) and 46.4% (Roemer’s approach) of inequalities stem from family « circumstances » and therefore constitute inequalities of opportunity. Effort accounts for only 6.1% (Roemer’s approach) to 8.1% (Swift’s approach). The remaining share of health inequality (approximately 50%) is due to demographic variables (gender, age).
Jusot, Tubeuf, and Trannoy conclude that while the distinctions between the approaches described above may play a role—with the share of legitimate inequalities rising from 6.1% to 8.1% (and even 11% in a control analysis)—the share of « circumstances » in health inequalities is much greater. Among these circumstances, social background, for example, would explain nearly 40% of health inequalities.
Conclusion
In France, the breakdown of equal opportunities is clear: income and health inequalities are largely determined by social background. While this fact is not particularly new, recent studies have managed to separate the sources of inequality (effort and circumstances) more systematically and isolate the role of social background more precisely.
The presidential election campaign, which is gathering pace in early 2017, offers the various candidates an opportunity to present their vision of a « fair society » and « social justice. » We can only hope that these candidates will take on board this research without distorting it.
References:
John E. Roemer, Theories of Distributive Justice, Harvard University Press, 1998.
http://www.hup.harvard.edu/catalog.php?isbn=9780674879201&content=reviews
John E. Roemer, Alain Trannoy, Equality of Opportunity, in Handbook of Income Distribution (chapter 4), edited by Anthony B. Atkinson and François Bourguignon, ScienceDirect, 2015.
http://www.sciencedirect.com/science/handbooks/15740056
Denis Magain, Post-Rawlsian theories of distributive justice. A review of the literature. Revue Economique, vol. 53, no. 2, 2002.
https://www.cairn.info/revue-economique-2002-2-page-165.htm
Jusot, Florence, Tubeuf, Sandy, and Trannoy, Alain, CIRCUMSTANCES AND EFFORTS: HOW IMPORTANT IS THEIR CORRELATION FOR THE MEASUREMENT OF INEQUALITY OF OPPORTUNITY IN HEALTH?, Health Economics, Vol. 22, issue 12, 2013.
http://onlinelibrary.wiley.com/doi/10.1002/hec.2896/abstract
Marc Fleurbaey and Vito Peragine, « Ex ante versus ex post equality of opportunity, » Economica 80: 118-130, 2013
http://onlinelibrary.wiley.com/doi/10.1111/j.1468-0335.2012.00941.x/abstract
Arnaud Lefranc, Nicolas Pistolesiet Alain Trannoy, Equality of opportunity and luck: Definitions and testable conditions, with an application to income in France, Journal of Public Economics, Volume 93, Issues 11–12, December 2009, Pages 1189–1207
http://www.sciencedirect.com/science/article/pii/S0047272709000905
Alain Tranoy, Equality of Opportunity: A progress report, Revue d’économie politique, 5 (Vol. 126), 2016.
https://www.cairn.info/resume.php?ID_ARTICLE=REDP_265_0621
John E. Roemer, Equality of Opportunity, Harvard University Press, 1998.
http://www.hup.harvard.edu/catalog.php?isbn=9780674004221
Swift, Adam (2005) Justice, luck, and the family: the intergenerational transmission of economic advantage from a normative perspective. In: Bowles, Samuel and Gintis, Herbert and Osbourne Groves, Melissa, (eds.) Unequal chances: family background and economic success. New York: Princeton, N.J.; Oxford: Russell Sage Foundation: Princeton University Press, pp. 256-276.
http://press.princeton.edu/titles/7838.html
John E. Roemer, Economic Development as Opportunity Equalization, World Bank Policy Research Working Paper, 2013.
http://documents.worldbank.org/curated/en/588981468251375061/pdf/WPS6530.pdf
Richard Anerson, Equality of Opportunity, Stanford Encyclopedia of Philosophy, 2015
https://plato.stanford.edu/entries/equal-opportunity/#FaiWorOffCriFEO
[1] John E. Roemer’sbook Theories of Distributive Justice (1998) presentsa more detailed account of the contributions of formalization to theories of justice.
[2] John E. Roemer and Alain Trannoy co-authored a chapter (Equality of Opportunity) in the Handbook of Income Distribution, on which this article is based. Alain Trannoy also wrote a « progress report » that presents the literature on equal opportunity in greater detail.
[3] Although already discussed by Rawls, Denis Maguain (2002) offers a review of the philosophical literature drawing on Rawls and post-Rawlsian theories.
[4] In this article, the English terms « efforts » and « circumstances » are retained.
[5] This definition of « circumstances » corresponds to that of Roemer. There are others that are slightly different (for example, Fleurbaey). Interested readers can find more details in Tranoy’s article (2016).
[6] In fact, several types of luck can be distinguished: pure luck and luck chosen by individuals.
[7] More precisely, Roemer proposes considering relative effort. Let us imagine that there are two « types » of individuals, those from a modest background and those from a wealthy background. Within these types, individuals could be ranked according to their « efforts. » Two individuals in the same rank (for example, the top 1%) should have the same result, regardless of their type.
[8] That is, the disposable income of a household normalized by the number of people in the household.
[9] They also discuss the consideration of pure luck in debates about equal opportunity.
[10] Again, in the case of education, an individual could not use a long education to justify a higher salary if their social background instilled in them a taste for academic effort.
