The degree of intergenerational income mobility is an important marker of economic equality in a society. On that front, Canada appears to be more equitable than the US (Corak 2004). Growing up in poverty in Canada is less limiting than in the US.
Given the large earnings premium associated with education (Card 2001, Heckman et al 2008), a person’s chances of moving from childhood poverty to adulthood prosperity depend heavily on the degree of educational mobility that exists.
While this mobility depends on many social and economic factors, governments can also influence it through student financial aid. Indeed, two recent surveys of the literature conclude that lowering the annual price (tuition less non-repayable aid) of post-secondary schooling by $1,000 leads to a 3 to 5 percentage point increase in attendance (Kane 2006, Deming and Dynarski 2009).
New research on aid and attendance
The structure of need-based aid almost certainly plays a major role in shaping the relationship between family income and post-secondary attendance. However, we know surprisingly little on the topic. In recent research (Belley et al 2011), we provide the first full accounting of financial aid and tuition as they relate to family income in both Canada and the US. We also examine the role of financial aid policy in shaping the US-Canada difference in the relationship between family income and post-secondary education attendance.
Average university tuition fees are about 40% higher in the US after accounting for purchasing power parity. This amounts to about 1,500 Canadian dollars per academic year. It is, therefore, tempting to assume that the cost of higher education is greater in the US. However, tuition (and other school) fees do not represent the full cost of attendance.
- First, students must forego labour market earnings while in school. A 2002 study by Burbidge, Magee, and Robb shows that foregone earnings are similar in Canada and the US, so this difference is not likely to contribute to differential post-secondary attendance patterns across the two countries.
- Second, non-repayable aid (grants) must be subtracted from tuition fees to yield a measure of net tuition, reflecting the effective price.
- Third, some students may be forced to live on a very limited budget while enrolled in school if they are unable to finance up-front costs and consumption through access to family resources or adequate credit. In this case, out-of-pocket costs (net tuition less available student loans) may become an important factor affecting enrolment decisions for credit-constrained youth.
A detailed analysis of net tuition and out-of-pocket costs suggests that the US is more generous at providing aid for the most economically disadvantaged, while Canada is slightly more generous towards middle-income youth. Figure 1 below suggests that net tuition at four-year public institutions is considerably lower in the US than in most of Canada for students from lower-income families. Figures for students living away from their parents are shown; however, the patterns are similar for students living at home. The Canadian figures are shown for Ontario (which is like most of Canada) and Québec, which has lower tuition fees and a stronger emphasis on non-repayable aid than other provinces. Lower-income students from the US and Québec generally face negative prices for post-secondary education, compared to about $2,000 for similar students in Ontario (and the rest of Canada). There is generally more parity as we move up the income distribution. Out-of-pocket costs are also lower for US students in the lower part of the income distribution (Figure 2). In this case, Québec is more similar to Ontario and the rest of Canada.
There are two primary reasons costs are lower for disadvantaged American youth, despite higher tuition levels. First, Canadian youth, even those from lower-income families, are expected to contribute to their own education. In contrast, lower-income youth from the US are generally exempt from contributing to their own education. Second, the US and Québec are more generous in terms of grants and scholarships at the lower end of the income spectrum. However, Québec is not so generous in terms of loan aid, which explains why out-of-pocket costs are about as high in that province as in the rest of Canada for lower-income youth.
Given the lower costs faced by disadvantaged American students compared with their Canadian counterparts, we might expect to see a weaker income–university attendance relationship in the US. This is not the case, however, as Figure 3 demonstrates with similar data in both countries (the 1997 cohort of the National Longitudinal of Youth in the US and cohort A of the Youth in Transition Survey in Canada). The gap in university attendance between youth in the highest and lowest parental income quartiles stands at about 20 percentage points in Canada. In the US, the gap is more than twice as large (roughly 45 percentage points). The difference between these income-attendance gaps is largely due to the lower attendance rate among low-income youth in the US. At the top end of the income distributions, attendance rates are about the same in both countries.
Why then do we see a much larger university-attendance gap in the US if costs are lower among the disadvantaged? Differences in family background, adolescent cognitive achievement, and local area of residence explain slightly more than half of the difference. Conditional on these factors, the gap falls to 18 percentage points in the US and to 7 percentage points in Canada. The conditional attendance gaps at four-year post-secondary institutions are about the same: 8 percentage points in Canada versus 17 percentage points in the US. A larger gap remains in the US even after accounting for many other observed characteristics.
It is conceivable that the greater generosity towards disadvantaged youth in the US is itself the result of very weak demand for post-secondary schooling among this group. In this case, the income-attendance gradient would be even steeper in the US relative to Canada in the absence of need-based aid. This can be observed in Figure 4, which reports counterfactual predictions about income-attendance gaps in the absence of need-based grants using standard estimates of the impact of net tuition on attendance.
Using a similar methodology, it is straightforward to ask how much aid would be required to completely eliminate conditional income–university attendance gaps. Figure 5 shows that the answer ranges between 60% and 100% for low-income students in both countries. Given higher levels of baseline aid, these figures imply considerably larger outlays in the US. Achieving equity through post-secondary schooling may be feasible; however, it would come at considerable cost.
Belley, P, M Frenette, and L Lochner (2011), “Post-secondary Attendance by Parental Income: Comparing the US and Canada”, NBER Working Paper No. 17218; also University of Western Ontario, Department of Economics, CIBC Working Paper 2010-3.
Card, D (2001), “Estimating the Returns to Schooling: Progress on Some Persistent Econometric Problems”, Econometrica, 69(5):1127-1160.
Corak, M (ed.) (2004), Generational Income Mobility in North America and Europe, Cambridge University Press.
Heckman, JJ, L Lochner, and P Todd (2008), “Earnings Functions and Rates of Return”, Journal of Human Capital, 2(1):1-31.