Mainstream economics typically emphasises efficiency over equity. One reason is because egalitarian notions can be subjective, and economists do not consider it their role to advocate how redistributive a society should be.

Notwithstanding, modern research does focus on what the effects of potential policies would be on quantitative measures of inequality (such as the Gini coefficient). By doing so, economists can evaluate whether certain redistributive policies can indeed achieve their stated goal, without passing any value judgment on the goal itself.

But such research is limited because there are several different notions of equality. Our study explores ways to quantitatively measure several notions of ‘equality of opportunity’ discussed by philosophers. Distinguishing such different notions is important: depending on which notion of equality is considered, some familiar real-world policy suggestions, such as increasing education subsidies, may have conflicting effects. While they are generally beneficial for the general populace under some notions, they can be harmful under others.

We show this in the context of an economic model in which successive generations of parents make decisions about their children’s education. Each generation of parents invest in their children’s education, and adult children decide whether or not to attend college. Middle-aged children receive financial support from their now elderly parents, while beginning to invest in their own children’s education. This model is estimated to the degree of inequality and intergenerational mobility observed in the United States.

The driver of inequality in the model is genetic ability (how good children are at learning at school) and whether parents are financially constrained when investing in their children’s education. Because education is so important, we find that eliminating other sources of subsidies and focusing them all into education can increase average welfare by 28 per cent, and decrease the Gini coefficient of lifetime earnings and wealth from 0.48 and 0.62 to 0.44 and 0.59, respectively.

But there is a moral dilemma around the impact on equality of opportunity. To understand why, consider a hard-working young adult who educates him/herself, attains a high-paying job, and is consequently able to afford high-quality education for his/her own children. Since these children are likely to achieve better economic outcomes than other children solely by the luck of having been born into a rich family, this may be considered inegalitarian.

On the other hand, if we thus force the parent to subsidise other children’s education at the expense of his/her own children’s, this may be considered unethical, especially if the parent had been motivated to work harder specifically to attain better education for his/her children.

Sign up for our newsletter here.

While we do not claim that either concept is ethically superior to the other, we do point out that such different notions call for different ways of measuring economic mobility if one is to maintain consistency.

Often, earnings or income is used as the basis for measuring economic inequality. When measuring mobility, however, we note that an ethical decision must be made on whether or not to deduct private education expenses from income.

First, consider the case in which we endorse that all children should have equal access to education, regardless of their parents’ efforts. If we want to measure the degree of inequality among a generation, then, we should not simply compare their income but first deduct the amount that they invest in their children.

To see this, consider two neighbours who earn exactly the same amount. Ms. Jones sends her child to public school and enjoys most of her income on other activities, while Mr. Smith spends most of his on private school tuition for his child. While it is likely that Mr. Smith’s child will grow up to be richer than Ms. Jones’, this comes at the expense of Mr. Smith himself, who lives meagerly compared to Ms. Jones. Instead, his efforts are rewarded by his child growing up to have better socioeconomic outcomes.

Thus without deducting the amount Mr. Smith spent on his child’s education from his income, we would be double accounting for his investments when measuring inequality: once in the parents’ generation and again in the child’s generation, when in fact such investments would only benefit the latter. That is, we would conclude that Mr. Smith is just as rich as Ms. Jones, while his child was richer than Ms. Jones’. While in fact, since we are taking the view that parents’ efforts should not be rewarded, Mr. Smith should be considered poorer than Ms. Jones — his income that went toward his child’s education did not benefit himself in any way.

Conversely, if parents’ efforts are to be rewarded, we need to compare children’s incomes conditional not on parents’ income, but lifetime wealth, that is, before deducting the amount they spend on their own children. Even though some of this wealth would go toward educating their children, we are taking the view that all their efforts -— including the efforts they undertook solely for the sake of their children — are benefiting the parents.

This distinction is important. If we believe in the view that parents’ efforts should be rewarded — in the sense that we deem it ethical that children who benefit from large i­nvestments from parents grow up to have better socioeconomic outcomes — equality of opportunity is almost achieved from focusing all government subsidies into education. In that case, parental conditions would determine only 2 per cent of inequality in their children’s generation. And while raising education subsidies would have the undesirable effect of rewarding parents who didn’t care to invest in their children, the effect would be minimal because the returns to their investment would be small.

But if parents’ efforts should not be rewarded — that is, we believe it is only ethical if all children have equal access to education regardless of their parents’ investments — we have to deal with the fact that parental conditions determine about half of children’s outcomes already (worse than if we were to only compare children’s earnings or income). Worse, even though equality of opportunity is so small, raising education subsidies barely help in our benchmark scenario, reducing the role of parental conditions by only about two percentage points (from 54 per cent).

For a significant change, much more drastically progressive policies would be called for, such as not providing any subsidies for the rich (which would be akin to preventing them from having access to all forms of public education) and fully subsidising the poor (which would need to include not only schooling but education that takes place at home).

If simply raising education subsidies does not help, what can a government do to expand equality of opportunity? And should society reward the efforts made by parents specifically to promote their children’s welfare, even if it generates larger inequality in subsequent generations? More research and collaboration are needed among social scientists across all fields to answer such questions.



  • This blog post is based on the authors’ paper Economic Policy and Equality of Opportunity, The Economic Journal, July 2018.
  • The post gives the views of its authors, not the position of the institutions they represent, the LSE Business Review or the London School of Economics.
  • Featured image credit: Photo by pixel2013, under a CC0 licence
  • When you leave a comment, you’re agreeing to our Comment Policy.

Tim Lee is assistant professor at the Toulouse School of Economics and a research affiliate in macroeconomics and growth at CEPR. His research interests are inequality, intergenerational mobility and human capital theory.



Ananth Seshadri is professor of economics and Todd E and Elizabeth H Warnock Chair in Economics at the University of Wisconsin, Madison. He has worked as a consultant at the State of Wisconsin Investment Board, a member of the board of editors of the Journal of Economic Literature, and a member of the executive committee of the Michigan Retirement Research Center.