Racial Differences in College Wage Premiums

Introduction: Understanding Racial Differences in College Wage Premiums

In recent years, the economics literature increasingly focuses on rising wage inequality in the labour market, with particular attention to the concept of the college wage premium. The college wage premium refers to the difference in average hourly earnings between individuals who complete only secondary education and those who graduate from university.

As college graduation rates rise, researchers seek to understand whether this trend is mirrored by changes in the wage premium, and crucially, whether these changes affect all racial groups equally. This analysis explores the evolution of college wage premiums in the United States, with a specific focus on racial differences, drawing on empirical evidence from the Integrated Public Use Microdata Series (IPUMS) for the years 1980 and 2014.

The Evolution of the College Wage Premium

Historical Trends and Definitions

The college wage premium began to climb sharply in the early 1980s. Evidence from James (2012) shows that between 1980 and 1990, college graduates earned on average 30% more than high school graduates. However, this premium began to contract after 1990, introducing new complexities for young people deciding whether to pursue higher education.

Technological Change and Demand for Skills

The expansion of technology across the economy has fundamentally altered the demand for skilled labour. Many roles previously accessible to high school graduates now require the technical expertise of college graduates. As a result, the supply of graduates must be matched by even greater demand to sustain or increase the wage premium. This interplay of supply and demand is central to understanding wage differentials.

Theoretical Perspectives on College Wage Premiums

Human Capital Theory

The human capital model, initially developed by Becker (1964) and refined by Mincer (1974), remains the cornerstone of wage analysis. According to this model, education and experience enhance an individual’s productivity and, consequently, their earnings. Mincer’s specification posits that the return to education can be precisely estimated if years of schooling are measured accurately and if each additional year of education yields a consistent increase in earnings.

Empirical Confirmation

Empirical studies consistently confirm that higher education and greater experience lead to higher wages. For example, Day and Eric (2002) found that high school graduates earned about one-third less than college graduates. The literature also documents fluctuations in the college wage premium, often linked to technological change and shifts in the supply and demand for graduates.

Fluctuations in the College Wage Premium

Technological Drivers

David (2014) attributes the rising college wage premium to increased use of advanced technologies in the workplace. Employers increasingly seek workers who can operate in computer-based environments, leading to a sustained demand for technically proficient graduates. As the structure of education evolves to produce more skilled workers, the wage premium is likely to persist or even grow.

Supply and Demand Dynamics

Murphy and Welch (1989) argue that the college wage premium is influenced by the relative growth of supply and demand for graduates. If the supply of graduates outpaces demand, the premium may fall. Conversely, when demand grows faster than supply, the premium rises. Their analysis, using real wage comparisons, found that from 1979 to 1986, real wages for high school graduates declined by 6.5%, while those for college graduates rose by 5.6%.

Racial Differences in the College Wage Premium

Early Evidence and Major Selection

Eide (1997) examined how changes in college wage premiums differ by race and found that differences in chosen majors account for much of the variation among men, but not women. This suggests that field of study may play a significant role in shaping wage outcomes, particularly for men who are not white.

Persistent Racial Wage Gaps

Economists note that people of colour (POC), especially Black and Latino individuals, tend to earn less than their white counterparts. This holds true even when controlling for education and other characteristics. This raises the question: do racial differences persist in the college wage premium itself, or does higher education level the playing field?

Data and Methodology

Data Sources and Sample Selection

This analysis uses data from the IPUMS, specifically the U.S. Census and American Community Survey. The study focuses on male respondents in 1980 and 2014 to avoid complications related to potential career interruptions among women for childcare. After data cleaning, the sample includes 196,692 observations for 1980 and 378,076 for 2014.

Key Variables

• Dependent Variable: Logarithm of hourly wage, derived from annual wage and salary income divided by total hours worked.
• Independent Variables: Education level (high school or college), potential experience (age minus years of education minus six), race (white or POC), geographical location, marital status, and family income.

Statistical Approach

The analysis employs Ordinary Least Squares (OLS) estimation, following the Mincer human capital model. Dummy variables distinguish between college and high school graduates, as well as between white and POC. Asians are excluded due to their disproportionately high educational attainment, which could bias the results.

Descriptive Statistics: Trends Over Time

Wage Growth and Educational Attainment

Between 1980 and 2014, the mean log hourly wage increased from $.69 to $2.72. The proportion of college graduates also rose, with college graduates exceeding high school graduates by 29.56% in 2014. The sample remains predominantly white, with whites accounting for over 89% of observations in both years.

Empirical Results: Racial Differences in College Wage Premiums

Mincer’s Model Findings

The results confirm that college education significantly boosts hourly wages. In 1980, college graduates earned 41% more than high school graduates; by 2014, this premium had risen to 70%. Experience also increases wages, though the effect diminishes over time, as indicated by the negative coefficient on the squared experience term. Whites earned 24% more than POC in 1980, a gap that increased to 26% by 2014.

Geographical and Socioeconomic Influences

Living in metropolitan areas positively affects wages, though the effect was stronger in 1980 (12.2%) than in 2014 (2.6%). Higher family income also correlates with higher hourly wages, reinforcing the importance of socioeconomic background.

Racial Differences in the College Wage Premium: Detailed Analysis

Comparing Whites and POC

The college wage premium is consistently higher for POC than for whites in both sample years. In 1980, POC college graduates earned 43.1% more than their high school-educated peers, compared to a 38.3% premium for whites. By 2014, the premium for POC had risen to 65.8%, while for whites it stood at 59.1%. Over the 34-year period, the premium increased by 22.7% for POC and 20.8% for whites.

Interpretation of Racial Wage Premium Differences

The higher college wage premium among POC may seem counterintuitive given persistent racial wage gaps. However, several factors contribute to this pattern. First, POC are more likely to attend low-quality high schools due to residential segregation and lower family incomes. As a result, POC high school graduates tend to earn less than white high school graduates. When POC do attend and complete college, the wage boost is more pronounced, reflecting the larger gap between their baseline earnings and those of college graduates.

Economic Interpretation: Supply, Demand, and Technological Change

Rising Demand for Skilled Labour

The increase in college wage premiums across all racial groups suggests that the demand for skilled, college-educated workers has grown faster than the supply. Technological advancements in the United States require a workforce capable of managing complex systems and adapting to new technologies. This dynamic drives up wages for college graduates, even as more individuals attain degrees.

Geographical and Socioeconomic Disparities

The wage premium’s magnitude is shaped by underlying inequalities in access to quality education and economic resources. Whites are more likely to live in affluent areas with high-quality schools, which boosts their earnings even at the high school level. POC, disproportionately represented in lower-income districts, face additional barriers to educational attainment and labour market success. Consequently, the wage gap between high school and college graduates is wider for POC, amplifying the college wage premium.

Policy Implications: Addressing Racial Disparities

Expanding Access to Higher Education

Given that high school graduates earn significantly less than college graduates, policies should aim to increase access to higher education, especially for underrepresented racial groups. Improving accessibility can help reduce wage disparities and promote economic mobility.

Improving Educational Quality in Segregated Areas

Targeted investments in educational infrastructure in segregated and low-income areas are essential to bridging the earnings gap. Enhancing the quality of schools attended by POC can help equalise opportunities and reduce the differential in college wage premiums across racial groups.

Balancing Supply and Demand

While increasing the number of college graduates is beneficial, policymakers must also consider the demand for skilled workers. If the supply of graduates exceeds demand, the college wage premium may decline. Therefore, efforts to align educational outcomes with labour market needs are crucial for sustaining the value of higher education.

Limitations and Future Research Directions

Endogeneity of Education

A key limitation of this analysis is the potential endogeneity of education in the wage equation. Individuals who choose to pursue higher education may differ systematically from those who do not, introducing bias into the estimated returns to education. While instrumental variable techniques can address this issue, suitable instruments are not available in the IPUMS dataset used here.

Exclusion of Women and Asians

The focus on males and the exclusion of Asians, due to data limitations and concerns about bias, means that the findings may not generalise to the entire population. Future research should seek to include a broader range of demographic groups to provide a more comprehensive picture of racial differences in college wage premiums.

Conclusion: Racial Differences in College Wage Premiums

Over the past three decades, the college wage premium has increased for both whites and POC in the United States. However, the premium is consistently higher for POC, reflecting deeper structural inequalities in access to quality education and economic resources. Technological change and rising demand for skilled labour have driven up wages for college graduates, but these benefits are not distributed equally across racial groups. Addressing these disparities requires targeted policy interventions to improve educational quality and accessibility, particularly in segregated and low-income areas.

Summary Table: College Wage Premiums by Race and Year

YEAR	White College Wage Premium	POC College Wage Premium
1980	38.3%	43.1%
2014	59.1%	65.8%

The table above summarises the key findings: while both groups see substantial gains from college education, POC experience a larger premium, largely due to lower baseline earnings among high school graduates.

Final Thoughts

Racial differences in college wage premiums highlight the complex interplay of education, labour market dynamics, and socioeconomic inequality. While higher education remains a powerful tool for economic advancement, its benefits are not distributed evenly. Policymakers and educators must work together to ensure that all individuals, regardless of race, have equal access to the opportunities and rewards that come with a college degree.

cause of this problem, economists have resorted to the use of instrumental variables to mitigate the biased estimates of the true return to education. Our paper does not employ the instrumental variable approach. The reason is that a good instrument is not available in the IPMUS data set.