A Model of Educational Investment, Social Concerns, and Inequality

• Published date: 01 October 2019
• DOI: http://doi.org/10.1111/sjoe.12299
• Date: 01 October 2019

Scand. J. of Economics 121(4), 1620–1646, 2019

DOI: 10.1111/sjoe.12299

A Model of Educational Investment, Social

Concerns, and Inequality*

Andrea Gallice

University of Turin, 10134 Torino,Italy

andrea.gallice@unito.it

Edoardo Grillo

Collegio Carlo Alberto, 10122 Torino, Italy

edoardo.grillo@carloalberto.org

Abstract

We consider a model in which educational investments entail productivity gains, signaling

power, and social status.The latter depends on the agent’s relative achievement in one of three

dimensions: innate skills, level of schooling, and income.We study the three scenarios separately

and characterize the conditions under which social concerns increase or decrease educational

and income inequality. Inequality increases (decreases) if education lowers the stigma suffered

by low-skilled workers less (more) than it boosts the prestige enjoyed by high-skilled workers.

We discuss the expectedresults of some policies in light of these findings.

Keywords: Education; signaling; social status

JEL classification:D03; D10; I20; I21

I. Introduction

The choice regarding which level of education to acquire is paramount in

an individual’s life. This decision not only shapes an agent’s job market

perspectives – and thus his expected career path and economic well-being

– but it also plays a fundamental role in determining an individual’s social

position. Sociological studies have long recognized education as one of the

main determinants of an agent’s social prestige and status attainment (Haller

and Portes, 1973; Hollingshead, 2011). In particular, it has been noted

that in many societies, low educational achievement can cause individuals

to suffer from social stigma (Solga, 2002), while high educational levels

convey social prestige (Archer et al., 2005).

*We thank participants in seminars and presentations at IMT Lucca, National University of

Singapore, SAET 2014 (Tokyo), EEA-ESEM 2014 (Toulouse), and ASSET 2014 (Aix-en-

Provence). A. Gallice is also affiliated with the Collegio Carlo Alberto. E. Grillo gratefully

acknowledges the Unicredit and Universities Foscolo Foundation for its generous financial

support through the Foscolo Europe fellowship.

C

The editors of The Scandinavian Journal of Economics 2018.

A. Gallice and E. Grillo 1621

Tab le 1 . Regression of self-reported social status on income and education

Country βInc βEdu

Australia 0.261 (0.024) 0.122 (0.020)

Brazil 0.194 (0.018) 0.055 (0.015)

China 0.443 (0.025) 0.061 (0.020)

Germany 0.394 (0.016) 0.137 (0.021)

India 0.084 (0.008) 0.075 (0.007)

Japan 0.160 (0.014) 0.053 (0.020)

Mexico 0.100 (0.012) 0.049 (0.015)

Morocco 0.607 (0.031) 0.167 (0.023)

The Netherlands 0.266 (0.031) 0.270 (0.023)

New Zealand 0.110 (0.025) 0.138 (0.044)

Nigeria 0.373 (0.022) 0.128 (0.019)

Russia 0.247 (0.019) 0.077 (0.017)

South Africa 0.232 (0.014) 0.164 (0.016)

Sweden 0.286 (0.026) 0.103 (0.022)

USA 0.380 (0.018) 0.115 (0.022)

Notes: Coefficients obtained from ordered probit regressions for each country of self-reported social status (measured

on a scale of 1 to 5, with 1 being the lowestclass and 5 the highest class) against income level, educational attainment,

and a set of controls. Controls are: gender, age, marital status, employment status, dummy variablefor religiosity,

dummy variable for whether the household saved money during last year, degree of financial satisfaction, and a

variable measuring whether the respondent’s job is manual or intellectual (see the Online Appendix for details).

Standard deviations are in parentheses.

Data source: WorldValues Survey Wave6, 2010–2014.

In principle, the correlation between an individual’s level of schooling

and his social standing can arise both directly, as a high education level

leads to social recognition in itself, and indirectly, because it is associated

with other key determinants of social status such as intelligence and income.

Using data from the World Values Survey,1Table 1 shows that, despite

substantial heterogeneity in the extent and relative strength of the two

relationships (probably reflecting socioeconomic and cultural differences

across countries), an individual’s self-reported social position is positively

associated with both his educational attainment and his income, even after

controlling for sociodemographic and economic characteristics.2

Starting with Veblen (1899), an extensive body of economics literature

testifies to the importance of status-seeking behavior in many economically

relevant settings; see Weiss and Fershtman (1998) for a survey, and see

1WorldValues Survey Wave 6, 2010–2014; see www.worldvaluessurvey.org/.

2Table 1 reports data from a small sample of countries from all continents. In Section IV of the

OnlineAppendix, we show that both relationships hold true (at the 95 percent confidence interval)

for 50 out of the 57 countries covered in theWorld ValuesSurvey and for which observations on

the variables used in the empirical analysis are available.

C

The editors of The Scandinavian Journal of Economics 2018.