Studying more pays-off, but how much?

We know that the more education we have, the higher our wages and the lower our chances of being unemployed. However, this is not always fully grasped by students and their families. This topic has long been studied by the economics of education. The first estimates of the relationship between education and wages were advanced by Mincer (1958) and, later on (1974), they were expanded based on the experience of individual workers in the labour market. Drawing on this relationship, Patrinos (2016) estimated that the wage returns of an additional year of education were approximately 10%. That is, the rates of return of an individual with an additional year of schooling would be, on average, 10% higher. However, these returns vary in two extents:

1. in terms of gender, with 11.5% higher returns for woman, compared to 9.1% for men, which is a common differential in all regions of the world;
2. and in terms of local economy, as developing country economies tend to have higher returns on education. Since education levels are usually lower in these economies, there is a greater appreciation of skilled workers in the labour market.

However, estimating this simple relationship between education and wages entails multiple challenges, and the main one is to distinguish between the innate abilities of an individual and those acquired by formal education. For example, the increased wage return of graduated workers may result not only from the knowledge accumulated during their additional years of schooling but also from their intrinsic cognitive abilities. In this respect, a well-known study (Ashenfelter & Krueger, 1994) compared the career path of twin brothers with very similar innate abilities, who grew up in the same family environment. In this case, the wage-education relationship was uniquely based on their years of schooling.

The portuguese case

The return of an additional year of education in Portugal, using data from the private-sector, is around 7.6% for women, and 6.4% for men (Campos & Reis, 2017). As shown in the figure below, this wage premium kept increasing until the 1990s, and remained relatively constant since then. It should be noted that, although a variation of one percentage point may seem low, we are looking at an increase based on one additional year of schooling. These figures reflect a change in the composition of the Portuguese workforce, with a higher participation of women in the labour market, a progressive increase of women's education levels compared to men, and a substantial increase in the number of workers with secondary and tertiary education.

Although an increase in the number of skilled workers could have led to a general wage premium reduction, this was not the case, as shown in the charts below. The increase in worker's qualifications was more than offset by the demand for qualified workers. If we look at wage premium variations per level of qualification (Figure 2), we see that, in 1986, the wage premium of an individual with higher education (compared to an individual with secondary education) was approximately 34%, for both men and women, and it continued to increase to 45% for women, and 50% for men. This development was positive until the mid-1990s, particularly for women, and remained relatively stable until 2009. Since then, it kept decreasing until 2013 (last analysed period, Campos & Reis, 2007).

Below we can see that, between 2009 and 2013, the wage premiums of graduated workers were lower for younger individuals. For example, in 2013, wage premiums were around 53% for 45-year-old workers, 50% for 35-year-old workers and 34% for 25-year-old workers.

In addition to estimating the average return per level of education, it is also important to distinguish wage premiums per professional area. Since there are no recent studies for Portugal in this regard, we will point out the results from studies carried out in the American (Altonji et al., 2012), English (Belfield et al., 2018) and Norwegian (Kirkeboen et al., 2016) labour markets. A common pattern was observed in these markets: tertiary degrees in STEM (Science, Technology, Engineering and Mathematics) courses yielded higher rates of return.

The measurement of return on investment in education presented here only considers the amount of education, and not the quality of the education acquired. Implicitly, this means that nine years of education, for example, would have the same impact for every individual, regardless of the education system.

A matter of quality, not just quantity

Hanushek and Woessmann (2011) present a simple exercise that illustrates how differences in wealth among countries are not only linked to the amount of acquired education but also to its quality. By establishing a relationship between the macroeconomic growth of a group of countries and the average number of schooling years of its workforce, we observed that the latter variable explained about 25% of the first. However, when we included the results of the PISA tests, per country, in this simple relationship, the years of schooling explained about 73% of a country's economic growth. This means that adding a "learning quality" metric significantly increases our explanatory power concerning the role of education on economic growth. Furthermore, when we considered individual wages instead of the aggregate economic growth, and added the results of the International Adult Literacy Survey (IALS) to Mincer's classic equation, we observed that these results had a significant impact on wage returns (Deny, et al. al. 2004). In the English case, for example, the quality of the attended university has a significant impact on labour market returns (Britton et al., 2016).
 

Beyond private gain – the social returns on education

The returns on education discussed above are the private returns; however, there can also be social returns. For example, if a person works on the same company or lives in the same geographical area as other highly educated workers, this could impact the person's productivity and lead to a wage increase, which would be a classic example of social returns on education. Furthermore, higher levels of education can also reduce crime rates (Lochner & Moretti 2004), and improve health behaviours as well as (Lleras-Muney 2005; Meghiret. Al 2012) the quality of civic and political participation (Milligan et al., 2004).

The fact that social returns on investment in education may differ from private returns is, in practice, of crucial importance. Most arguments in favour of public education point out that education not only favours those who receive it but also generates benefits that are shared by society at large.

Therefore, the magnitude of the social return on education is a crucial measure for assessing the efficiency of public investment in this area.

The high returns on higher education indicate that schooling continues to be a profitable private investment. In a context of budgetary constraints, policymakers are challenged to ensure the quality of the public education system and, at the same time, create the necessary conditions to allow lower-income students to have access to higher education.
When making decisions, policymakers need to bear in mind that investing in the earlier stages of education will increase the returns in the later stages (Cunha et al., 2010). Therefore, investments in tertiary education must not be made at the expense of primary and secondary education. These tensions need to be taken into account when defining and allocating resources and funding to public education. This matter is highly relevant for current debates on higher education financing, especially those concerning tuition fees and access to study grants and loans, since it is not clear whether, at this level, the private gains will be accompanied by clear social gains.