September 24, 2013

The growing economic clout of the college educated

Last week’s data released by the Census Bureau vividly demonstrated the growing income inequality in the U.S. But deeper in the data, there’s another related finding that is also striking: The growing income accruing to the nation’s college-educated households.

FT_13.09.23_CollegeEducated_1_420pxFor the first time on record, households headed by someone with at least a bachelor’s degree received nearly a majority (49.7%) of aggregate U.S. household income; nearly one out of every two dollars went to the college educated.  In 2012 one-in-three households was college educated, so, put another way, half of the aggregate U.S. income goes to one third of the households.

In 1991 (the earliest year comparable figures are available) college-educated households only received 37% of the nation’s aggregate income. In 1991 about one-quarter of households (23%) were college educated.

The share of the income pie received by households with only a high school education or less fell 15 percentage points from 1991 to 2012.  The share of household income going to households with some college (including those with an associate’s degree) increased modestly over the same period (23% to 25%).

Since educational attainment has risen and there are more college-educated households, one would expect the college educated to receive a growing share of the pie.

FT_13.09.23_CollegeEducated_2_420pxBut the data clearly indicate that the growing economic fortunes of the college educated go beyond sheer numbers. College-educated households are the only households whose incomes have grown on a per household basis from 1991 to 2012.  Household income increased 9% (from $92,289 to $100,637) for those whose highest education was a bachelor’s degree.  Incomes were up 20% for households with professional degrees.  In contrast, household incomes have declined for households who do not have at least a bachelor’s degree.

Before breaking down the nitty-gritty of the college-educated households’ income gain, it should be noted that a number of factors are likely at play in boosting the household incomes of the college educated relative to less-educated households.  A primary factor is the better fortunes of the college educated in the labor market. The Georgetown Center on Education and the Workforce finds that college graduates earn nearly twice as much as workers with just a high school diploma.

But the household income differences between the college educated and lesser educated go beyond the labor market.  College-educated households are more likely to be married and thus more likely to have secondary earners contributing to household income.

In addition, my research on “assortative mating” or “who marries whom” shows that married college-educated persons are more likely to have a college-educated spouse. Thus, they are more likely to have a spouse with high earnings. For example, in 2011, 75% of married men ages 30 to 44 who are college educated also have a college-educated wife.  Among their married counterparts with a high school education, only 17% have a college-educated wife.

Between 1991 and 2012, the aggregate household income of college-educated households increased by $2.1 trillion according to the Census data.  Over the same period, the share of all households who are college educated increased from 23% to 33%.  How much of the $2.1 trillion income gain received by the college educated is due to growth in numbers versus growth in income per college-educated household?  If the fraction of households who are college educated had remained constant at 23%, instead of rising to 33%, the income pie going to the college educated would only have grown by $0.8 trillion.  So, over half of the income gain of the college educated is due growth in numbers. But a substantial portion reflects their improving income fortunes.

Topics: College, Educational Attainment, Income, Income Inequality, National Economy

  1. Photo of Richard Fry

    is a senior researcher focusing on economics and education at Pew Research Center.

11 responses to “The growing economic clout of the college educated”

  1. Is there any data regarding the cost of the college educations? My wife & I both have associates degrees which we funded via scholarships & working. Our household income is significantly higher than the figures stated here so we may be an exception to the stats here. I can’t help but wonder what the “net” effect on income is when considering the cost of higher education. I hear of so many folks with prestigious degrees & great incomes but they also have great debt as a result of school loans, some of which they will be paying for the rest of their lives. That’s more of a money management issue really but more income is not necessarily a blessing if it costs so much to get the income that the net effect is a wash. I would be interested in seeing that data if it is available.

    • Richard Fry says:

      Good question and it comes up often. First, an observation regarding your household income. The figures provided are national averages. For any average, there will be households above average. Fortunately for you, your wife & you are above average.

      As far as the cost of the education netted against the income gain, economists do those sorts of calculations. For example, take a look at Figure 1.3 of the College Board publication called Education Pays. Here is the link:

      In the case of an assoc.s degree, they suggest that given the typical cost of an assoc degree and assuming a 6.8% interest rate, that the net earnings of an assoc degree holder will overtake a high school graduate by age 33 or 13 years of work.

      Thanks for your interest.

  2. Jose says:

    Could you please tell why had the fraction of household college educated remained constant their income had grown 0.8 trillion?

  3. Geoff says:

    Did you further break down the data by income group? For example, what were the increases in income for those with college degrees (Bachelor or higher) by income quintile? My guess is that those with college degrees in the lowest income quintiles saw very modest gains, while those with college degrees in the top income quintile (or the top 10%, 5%, 1%) saw the lion’s share of the income gains.

    • Richard Fry says:

      No, I was not able to break down the incomes gains by education further by income group. But, I suspect that your supposition is correct. Within education groups it is likely that income increases were not uniform but more limited to relatively higher income subgroups. This can be tabulated, it is just not available in the historical income tables that the Census Bureau publishes.

  4. I’d be interested to see this data normalized by size of each cohort. Does the college-educated “share” rise primarily due to the growth in the size of the cohort or is the shift more correlated to an increasing dearth of living wage jobs for those without college educations?

    Unlike CEO pay multipliers in the news of late, I would find it hard to vilify moving 10% of the population a rung up the ladder in the last 20+ years.

    • Richard Fry says:

      See the last paragraph of the post. If I understand your query, you recognize that part of the reason that college-educated households receive a growing share of national income is because a growing portion of households are college-educated. Clearly that is an important factor. Over the past 20 years the share of households headed by a college-educated person has risen from 1-in-4 to 1-in-3. So college-educated households get more income in part because there are more of them. But as the last paragraph reports, growth in numbers is only part of the story. Keeping their numbers constant, college-educated households are the only households with rising incomes (on an average basis), so they are receiving a rising share above and beyond their growing numbers.

  5. Hanan Harchol says:

    Excellent article! I was wondering if you might have this data broken down by state? ( I am a New York City high school teacher and am compiling data for a presentation to my students on the importance of earning a college degree) Thank you!

    • Richard Fry says:

      Thanks, I am glad that you found it useful. Unfortunately, I don’t have state tabulations. It could be done off other data sets, but the Current Population Survey (the basis for this blog) is not especially well-suited for more detailed levels of geography.

  6. Willie says:

    1. State the confidence interval in two ways:

    a. The lower limit to the upper limit

    2. b. As the sample proportion +/- the error term

    Answer these based on the research Pew performed on The rising cost of NOT going to college article.


  7. Willie says:

    Richard, oftentimes, surveys use generic terms in their research such as men between the ages of… or women between the ages of…

    Does Pew’s research include men and women are just or white males only?

    Thanks, again

Leave a Reply

Your email address will not be published.