The Effect of Education on Income in IL Essay Example

to make students a lot of money in the real world. Despite the years of education required, a PhD in medieval studies is unlikely to make a seven-figure income.

While there are rare, very talented people such as Lawrence, Gates and Zuckerberg who are able to forge their own lucrative paths with their talent instead of diplomas, we wanted to see what the data says about the impact of education levels on income. Education level is “often referred to as an investment in human capital,” according to the St. Louis Federal Reserve Bank. Bachelor’s, master’s, professional and doctorate degrees open the door to many skilled job opportunities that don’t even consider applicants without a college degree.

The goal of our research project is to examine if there is evidence that a person’s level of education has an effect on their income. Our data was taken from a survey of Illinois residents between ages 18-65 in 2017 (approximately 78,617 observations) conducted by the U.S. Census Bureau as part of the American Community Survey.

Statement & Explanation of Hypothesis

Our hypothesis is that education level has a positive impact on an individual’s income, holding other variables constant, because according to economic theory, a higher education level increases human capital and productivity and that leads to a higher income. We wanted to verify that there is a correlation between education to income, while determining the extent to which education has an effect on income.

However, we tested some other factors that we thought might have some collinearity with education and could impact income, such as gender, age, having a disability and English proficiency. Having a high level of education does not necessarily cause

an individual to have a high income, so it is not a causation, but we believe there is a direct correlation between income and education.

Literature Review

The effect of education on income is a topic that has generated plenty of studies and discussion in the U.S. for decades so there is no lack of literature on the subject. American workers who have graduated from college, including those who have a further education, are paid 74 percent more on average than workers with only a high school education, according to a statistic from the Organization for Economic Cooperation and Development cited in a 2014 New York Times article.

While we are not studying the differences in income, it is a staggering statistic. The extra money earned by workers with a college degree is often referred to as the “college wage premium,” according to the St. Louis Federal Reserve Bank.

Government policy is one of the key reasons that the correlation between education and income has been studied so extensively. Income inequality is a hot topic in the national dialogue. In fact it is the biggest economic challenge facing the U.S. over the next ten years, Federal Reserve chairman Jerome Powell said at a town hall meeting on February 6, 2019. “The U.S. lags now in mobility. And that’s not our self-image as a country, nor is it where we want to be,” Powell said, according to The Washington Post. “We have some work to do to make sure that the prosperity we do achieve is widely spread.”

There are many different opinions on what can solve income inequality, such as whether increasing educational funding in impoverished areas will lead to

more of the students eventually graduating from college, raising the minimum wage and increasing taxes on higher earners. We are not trying to solve that massive issue in this study but the literature out there, and hopefully our own conclusions, can offer some insight into the effect education really does have on income.

The Federal Reserve Bank of St. Louis has extensively studied, researched and written about the effect of education on income. They published an essay series called, “The Demographics of Wealth: How Age, Education and Race Separate Thrivers from Strugglers in Today’s Economy” in 2015 using data from the Federal Reserve Bank’s Survey of Consumer Finances with information from over 40,000 families spanning from 1989 through 2013. They only chose to study the education levels and incomes of heads of households at least 40 years old because most adults have completed their education by 40.

In the research, the St. Louis Fed found a “strong correlation” between education level and income and concluded that “the connections between education and wealth are likely to become even stronger in the future.” But the authors note that the connection between money and education is a just a correlation, not a causation because it is not guaranteed that a higher level of education will result in a higher income.

They cite many other factors that have an effect on income: natural ability, family background, incentive to become financially savvy, inheritances and better health and/or longer lifespans. It was noted that individuals who have more education are more likely to benefit from already having one (or more) of those factors but they are not caused by education.

Not only did they find

that education impacts income, they found that the median income actually decreased (when adjusted for inflation) for workers without a high school diploma, a high school diploma, an associate’s degree and a bachelor’s degree from 2013 to 1989.

Individuals with more education also had a better personal finance situation with “more liquidity, a better mix of investments and lower leverage” than those with lower levels of education. They are more likely to accumulate wealth over time because of practices such as “regular saving habits, timely payment of all obligations and conservative financial practices, such as holding adequate cash reserves, investing in a broad array of assets and borrowing moderately,” the research found.

The U.S. Department of Labor’s Bureau of Labor Statistics put together a graph using their extensive data to show the relationship between education level, unemployment rates and median usual weekly earnings. Their data found that individuals with a professional degree have the highest median usual weekly earnings at $1,836 and an unemployment rate of 1.5 percent.

It is a staggering difference compared to individuals with less than a high school diploma who have the lowest median usual weekly earnings at $520 and an unemployment rate of 6.5 percent.

Data

The data used in our analysis was from the Public Use Microdata Sample files obtained from the American Community Survey. The American Community Survey is a survey conducted by the US Census Bureau that takes data from a sample of households across the country and looks at various variables including a person’s age, weight, marital status, employment status, etc. The data used was taken from the 2017 year’s data and we decided to only use data from the state

of Illinois.

We chose to study the data from one state because it offered so many data points (78,617 people) and a localized estimate would give us more accurate results because education and income can vary across the country.

We also chose to study the 2017 data which is the most recent available to make it timely and relevant. We classified education level into these different categories: non-high school graduate or equivalent (GED), high school graduate or equivalent (GED), associate’s degree, bachelor’s degree, master’s degree, professional degree (such as a law degree or MBA) and a doctorate degree.

To complete our analysis, different variables that we believed would influence a person's income were included. With a person’s income acting as the dependent variable, there are plenty of factors which could possibly affect a person’s income so obviously not all were included. Some of the relevant factors considered were a person’s years of education, age, gender, English proficiency, and if a person is disabled.

We made the main focus on a person’s years of education as we believed that that would be one of the more impactful variables on a person’s income. However, the original data on education was categorical by degree, some liberties were taken and the degrees were converted into years of education by taking the average amount of years it takes in order to complete said degree; so 2 years for an associate’s degree, 4 years for a bachelor's, etc.

Another variable that we looked at was gender because we wanted to analyze who had the highest income with their educational attainment. For the purposes of this paper we are making the assumption that most of these people

spent their time out of school working and building up experience.

Age was also an important factor due to its relation with income and education; for this variable ages below 18 and greater than 65 were excluded as we took the assumption that most people below 18 were either not working, only had small part time or seasonal jobs due to still being in high school, and many people over 65 have retired.

English proficiency was also looked at to see how the regression changed accounting for people who are not fluent in English due to the fact that it may be more difficult for them to find steady work. The final independent variable we looked at showed whether or not a person was disabled, as we believe that a person’s disability could easily have an effect on their potential income.

Descriptive Data

The population distribution includes the incomes of individuals in the State of Illinois that satisfy the following criteria: (1) aged 18-65; (2) level of education (3) proficient in English (4) gender; and (5) not disabled. The dependent variable is income and the independent variables are as follows: education, age, gender, disability, and English language skills.

In our cross-sectional data set, gender and disability are binary variables. The age ranges we used in our analysis was 18 to 65 years old; the average age being 42 with a standard deviation of 14. In the analysis of the dependent variable, annual income, the mean was $46,154, with a standard deviation of $66,347. The income reported in the data set ranges from -$6,100 to $1,149,500.

In our next variable, school, we look at the education levels of the reported of the

18-65 year olds. The average education level of this group is 18.4 years which translates to “Some college, but less than 1 year”; with a standard deviation of 3.3.

Lastly, we observed individuals who speak English as a second language, and their proficiency levels. We noted 15,572 observations, which is about 20% of the data set. The four levels of English proficiency individuals can report were (1) very well (2) well (3) not well (4) not at all. The average result was, 1.58, which falls in between “very well” and “well” and, the standard deviation was .85.

When comparing the independent binomial variables, disabilities and gender, with the dependent variable, income, it is revealed that income difference is similar. The differences in income for “not disabled” and “disabled” is approximately $25,303; whereas the difference in income between males and females is approximately $22,528. This result is surprising, because we did not anticipate the disparity to be as equal.

Empirical Tests

Since the data set that was used for the analysis contained cross sectional data, multiple linear regression was conducted across the data for different combinations of population variables. From the distribution curves of income versus the curve for the natural log of income, it was determined that the natural log of income was a better fit for the analysis since the histogram displayed a bell curve that looked close to being normally distributed.

Level of education was broken down into a subset of dummy variables with no attainment of a high school education or GED as the base group. English proficiency was also broken down into dummy variables with the base group containing persons who are fluent or native English

speakers. Gender was included as a dummy variable with female as the base group and disability was included as a dummy variable as well with the base group containing individuals without any disabilities.

The final independent variable was age and this variable was tested as both a standard variable and also as a set of dummy variables with the base group containing 18 year old individuals. The age dummy variable group starts with the value useage1 for 18 year old individuals and goes up to useage48 for 65 year individuals and increases on 1 year increments.

The first test that was conducted was the regression using the natural log of income as the dependent variable with age, level of education, english proficiency, gender, and the presence of a disability as the independent variables. After running this regression, a BP-Test was performed on the result and it was seen that the data contains some heteroskedasticity. Since this was found in the results, the remaining regressions were conducted using robust standard errors.

The next regression that was run was the same regression that was previously run, but including robust standard errors (Appendix A.1). Subsequently, an F-test was run on all of the independent variables other than age with the null hypothesis being that the result would be equal to zero meaning that each of the variables being tested were jointly insignificant.

The results of the F-Test over these 10 variables was very large (1447) and greater than the critical value (2.32 at 99% significance), thus leading to the rejection of the null hypothesis, meaning that the combination of these variables is jointly significant.

The final set of regressions that were conducted included

a multiple linear regression of each of the independent variables, or subset of dummy variables for each variable, against the natural log of income to identify the interaction and significance in the value of the adjusted R-squared for each of these variables (Appendix A.1).

Summary of Findings

The multiple linear regression containing all of the independent variables, including age as a subset of dummy variables, was determined to be the best fit for the data that was given and in the testing of the overall hypothesis. Every variable included in the analysis was significant at the 1% level of significance and the overall regression had an adjusted R-squared value 32.5%.

It was determined to use age as a dummy variable because in the regression that included age as a dummy variable, it was seen that after a certain amount of time, an increase in age had a diminishing effect until at a certain point, around 51 years old, where each additional year had a reduction of income.

This could be due to retirement for high earning individuals skewing the data towards lower income individuals remaining in the workplace. When compared to the regression using age as a standard variable, it was seen that the beta value for age was positive and was a constant for each additional year. Overall, it was determined that age as a set of dummy variables was a better fit based on the adjusted R-squared and the ability to have a more detailed representation of the effect on age versus income.

From the beta values calculated in the regression, it can be seen that at its highest, at 51 years of age with a beta value

of 2.24, that a 51 year old individual will make 224% more than an 18 year old individual. For education, someone with a bachelors will make 92% more than someone without a high school education, and a masters degree will earn 113% more than someone without a high school education.

The highest education earners, those with a professional degree, will earn 150% more than someone without a high school education. Someone who is not fluent in English, but speaks English well will tend to earn 5.8% less than fluent English speakers and someone who speaks English poorly will earn 22.7% less than fluent speakers. Males will earn 47.7% more on average and individuals with disabilities will earn 63.4% less than individuals without disabilities.

For each of the multiple linear regressions based on the different variables being tested (age, education, English fluency, gender and disabilities), all of the variables were still significant at the 1% level of significance. The adjusted R-squared value for the age regression was the highest at 18.3% followed by education at 13.7%, disability at 2.8%, gender at 2.7% and English fluency at 0.3%.

Conclusion

In debating whether or not a person’s education would have an effect on their yearly income, we found the results to clearly dictate the significance effect education has on income. Our analysis shows that someone with a bachelors will make 92% more than someone without a high school education, and a masters degree will earn 113% more than someone without a high school education and those with a professional degree, will earn 150% more than someone without a high school education.

We realised from our regression how age also played an important role

on income and this helps confirm how people in the within the age rage 18 - 65 generate most income. These help confirm many common and well known assumptions that are seen in today’s society. Another common assumption that is a major talking point in many circles is the accusation of female income being lower than male.

This can be explained by the fact that “men take 48% fewer unpaid hours off and work 83% more overtime hours per year than women. The reason for these differences is not that men and women face different choice sets in this job. Rather, it is that women have greater demand for workplace flexibility and lower demand for overtime work hours than men”. (Parker et al., 2015; Bertrand et al., 2015).

Although, even with all these factors a person’s education still seems to have the most significant effect on income. This relationship does lead to one major issue, the problem of opportunity cost. Even though one could earn over $8000 by going to school for another year, that is potentially one year without income, or working only part time.

There then arises the cost of schooling where people will spend well over a eight thousand dollars to go to school for one more year. This dilemma can cause many issues both on the personal level, and the federal level when debating what the best choice to make is, choose one more year of work, or one more year of school.

References

1. Boshara, Ray, et al. “The Demographics of Wealth: How Age, Education and Race Separate Thrivers from Strugglers in Today’s Economy.” Economic Research - Federal Reserve Bank of St. Louis, Federal Reserve
   

Bank of St. Louis, May 2015, www.stlouisfed.org/~/media/files/pdfs/hfs/essays/hfs-essay-2-2015-education-and-wealth.pdf.