Americans Under-Educated

A college degree is becoming a virtual necessity to making a good living in America, yet most still aren't pursuing higher education.

In a series of posts on demographics over at The Glittering Eye (“The Breakdown,” “The Young Aren’t Getting Enough Education,” and “Education Is More Necessary Than Ever” thus far) Dave Schuler makes the case that we’re looking at the current economic downturn incorrectly.

He presents a chart showing that only 29.6% of Americans in the 25 to 34 cohort have a bachelor’s degree despite clear evidence that education and income are inextricably and increasingly linked.

So, Dave is in a quandary:

The disparity is getting larger rather than than staying the same. Unfortunately, young people, particularly young men, aren’t seeking college or advanced degrees in sufficient numbers.

I do feel the need to register some skepticism about the overall significance of post-graduate education. I think that the very high earnings that can be attributed to some of those with post-graduate degrees notably physicians and MBAs or even the extremely high earnings of a small subset within those obtaining post-graduate degrees, e.g. attorneys, skews the entire result.

Not only that, but it’s not at all clear that the trend is sustainable.  Would, say, doubling the number of people with degrees in PhysEd, Criminal Justice, and Elizabethan Poetry really result in a major increase in income?  Or is it simply a combination of certain career fields needing advanced degrees as an entry point and the fact that people who are likely to excel in the workplace are also the kind of people who make the right choices early and thus get a degree?

Certainly, the answer isn’t to stop with a high school diploma.  But it’s not entirely clear that higher education is the solution to all that ails us, either.

FILED UNDER: Uncategorized, , ,
James Joyner
About James Joyner
James Joyner is Professor and Department Head of Security Studies at Marine Corps University's Command and Staff College. He's a former Army officer and Desert Storm veteran. Views expressed here are his own. Follow James on Twitter @DrJJoyner.


  1. Juneau: says:

    Not to indulge my normal, facetious self,  but why worry about the college education level when we appear to be perfectly willing to accept a 50% graduation rate from our high schools?  The problem starts at a much younger age and it directly relates to the ( lack of ) competence and passion in our educators.  The teachers unions and the NEA are all about themselves, and the truly good teachers, of which there are many, have no financial incentive or reward for their excellence.
    Root causes, root causes…

  2. Boltwan says:

    I would suggest that people make a semi-rational choice as to what education they need for the field they choose to work in.  Hence, the correlation between vo-tech degrees/associates and income also.  Also, perhaps people are more cognizant of the ‘bang for the buck’ they are or are not getting from continuing their education.  People can choose to go to college for a few extra years or get the education they need to get into the workforce and build work experience(and their paycheck) to the point that they are making an equivalent income compared to people entering the workforce with a more advanced degree.

  3. Steve Plunk says:

    The specter of high debt and possible unemployment at graduation can scare away many potential students.  Higher education is overpriced regardless of pay differentials.  As with health care we spend too much time working on how to pay for it rather than how to control the costs.  If society wants more educated citizens we better get busy on the cost control issues.

  4. john personna says:

    When you are being normative (making recommendations for abstract high school graduates) it is easy to say “get a degree.”  In aggregate it is true.  For specific individuals though, that might not be enough, or be the wrong course.  It depends on the person, and it depends on the degree.
    At the top level though, sure.  Getting a degree is more important than ever, but also less a guarantee than ever.
    Just curious though, how would you plot changes in college costs over those same years?  How would a scatter plot with each year’s ROI look?  I bet it would be scary.  Recent years would show much higher costs without the level of opportunities that 1980s or 1990s graduates enjoyed.
    Sure, on average get a degree … but don’t finance a degree that doesn’t have solid market potential.  Getting an unmarketable degree won’t help you or spur the economy.  In some cases “a plan” might be better than a degree.

  5. TangoMan says:

    <I>He presents a chart showing that only 29.6% of Americans in the 25 to 34 cohort have a bachelor’s degree despite clear evidence that education and income are inextricably and increasingly linked.</i>
    I know that Dave used to read GNXP so I’m surprised that he’s focusing on the education – income correlation without mentioning the role of intelligence. There is a rich literature which addresses this issue.
    <blockquote>Rather than simply describing the need to control for measured cognitive ability, other studies argued for the additional need to include ability-education interactions. These studies note that the receipt of an additional year of schooling may have differential impacts across individuals, and this variation can potentially be explained through differences in individual ability. For example, Hause (1972), Blackburn and Neumark (1993), Murnane, Levy and Willett (1995), Grogger and Eide (1995), Heckman and Vytlacil (2001), and DiNardo and Tobias (2001) and Tobias (2003) investigate the question of whether returns to schooling vary with ability. Generally, these studies find that higher ability individuals benefit more (in terms of wages) from additional years of education than lower ability individuals.</blockquote>
    <a href=””>The Rising College Premium in the Eighties: Return to College or Return to Unobserved Ability?</a>
    <blockquote>The changes in the distribution of earnings during the 1980s have been studied extensively. The two most striking characteristics of the decade are (a) a large increase in the college/high school wage gap, and (b) a substantial rise in the variance of wage residuals. While this second phenomenon is typically implicitly attributed to an increase in the demand for unobserved skill, most work in this area fails to acknowledge that this same increase in demand for unobserved skill could drive the evolution of the measured college premium. In its simplest form, if higher ability individuals are more likely to attend college, then the increase in the college wage premium may be due to a increase in the relative demand for high ability workers rather than an increase in the demand for skills accumulated in college. This paper develops and estimates a dynamic programming selection model in order to investigate the plausibility of this explanation. The results are highly suggestive that an increase in the demand for unobserved ability could play a major role in the growing college premium.</blockquote>
    The relationship between schooling and income was actually empirically tested in Hong Kong and the results didn’t support a strong causal relationship.
    <a href=””>Shrinking Earnings Premium for University Graduates in Hong Kong: The Effect of Quantity or Quality?</a>
    <blockquote>In 1989, the Hong Kong government embarked on a program to increase the provision of first-year first-degree places. The expansion of tertiary education represents a large and exogenous increase in the supply of university graduates to the territory. This article measures the labor market effects of the expansion program by studying the changes in earnings premium for university graduates. Two alternative hypotheses-crowding and quality effects-are identified to explain why the earnings premium shrank. The results support the view that the declining quality of university graduates is the prime candidate for the declining earnings premium.</blockquote>
    Here is an interesting study which look at HS class rank longitudinal data.
    <a href=””>Ability, Educational Ranks, and Labor Market Trends: The Effects of Shifts in the Skill Composition of Educational Groups</a>
    <blockquote>This literature has tended to ignore changes in the composition of educational groups resulting from the dramatic increases in educational attainment over the past few decades. One particularly extreme example comes from Autor, Katz, and Krueger (1998) which compare the wages of college and high school graduates between 1940 and 1996. In 1940, college graduates made up the upper 6.4 percentiles of the educational attainment distribution, while high school graduates were between the 68th and 87th percentiles (p.1174). By 1996, college graduates made up the upper 28 percentiles, while high school graduates were between the 9th and 43rd percentiles (p. 1174). If educational attainment and ability (or unobserved skill) are positively correlated, then comparisons such as these require heroic assumptions in order to be interpretable as changes in only the return to skill. Given that the decline in educational rankings for high school graduates has fallen much faster for high school graduates than it has for college graduates (due to the expansion of the some college group), one might expect that changes in the composition of educational groups would have resulted in a large increase in the college/high school wage differential between 1940 and 1996, even in the absence of any change in the returns to ability and schooling. In this light, the six percentage point increase in the college/high school wage differential between 1940 and 1996 (p. 1174) likely indicates that wage inequality decreased rather than increased between 1940 and 1996.</blockquote>
    Dave notes that the slope of the trend lines indicates an interesting story. I think his observation is on the money. I think that what we’re seeing is that there are increasing returns to intelligence and that education is merely a proxy variable for IQ. As noted above, the composition of the educational cohorts is changing across time in that the demarcation lines between the groups get pushed ever lower into the class ranks, as measured by percentiles.

    Now, the Hong Kong experience suggests that the trend lines would be even steeper is the quality of the students that comprise each ranking wasn’t decreasing across time.

    The core issue here is value creation and the allocation of that value to the worker. How much value can an average high school drop-out create compared to the value created by an electrical engineer and what are the supply-demand labor situations for these two worker groups and how much bargaining power can they exercise to capture an increasing portion of the value they create in the labor marketplace.
    James hits an interesting point by questioning the economic value that can be created by training in Elizabethan poetry. So, if this Elizabethan poetry student goes on to earn a substantial income is it BECAUSE of the training received in university or is it because of the training or skills that are separate from her university experience?
    The upshot, I believe, is that employers will find suitable employees regardless of educational background with the proviso that the particulars of the job don’t require skill sets taught in an academic discipline.

  6. Brett says:

    You know what I’d like to see? A breakdown of the college wage premium by degree area. I have an unfounded suspicion that much of the wage premium is being driven by a number of degree areas, with many others (particularly in the humanities) not offering much of a premium at all.

  7. TangoMan says:

    <I>You know what I’d like to see? A breakdown of the college wage premium by degree area. I have an unfounded suspicion that much of the wage premium is being driven by a number of degree areas, with many others (particularly in the humanities) not offering much of a premium at all.</i>
    I can offer research somewhat related to your request.
    <a href=””>Returns to Graduate and Professional Education: The Roles of Mathematical and Verbal Skills by Major</a>

    <blockquote>Murnane, Willett and Levy (1995) found that rising returns to mathematics skills can explain a substantial fraction of the observed increase in returns to college between 1978 and 1986. The effect was stronger for women than for men. Grogger and Eide (1995) and Levine and Zimmerman (1995) also reported that standardized mathematics scores or having taken more mathematics classes had a significant positive impact on women’s wages but not men’s wages.

    <blockquote>Our review of the literature that concentrated on lower levels of education suggests that two effects are potentially at work: 1) Rising returns to cognitive skills may have increased the opportunity costs of attending graduate school, limiting incentives to pursue post-graduate education in the areas where the returns are rising the most rapidly. Consequently, the most able students opt not to pursue graduate education in favor of capturing returns to those skills in jobs they can acquire with a bachelor’s degree. 2) The marginal product of cognitive skills may have risen atypically in post-graduate training, raising the returns to graduate training relative to lower education levels.

    These two possibilities would have opposite effects on incentives to attend graduate school and on observed wages.</blockquote>

    <blockquote>If major-specific skills at the bachelor’s degree level are increasing in market value, then they will tend to lower incentives to pursue graduate work in that field. Conversely, majors whose skills are falling in value at the bachelor’s level will have disproportionately high numbers of graduate students. If this sorting effect drives lower earning bachelor’s degree recipients into graduate school and drives higher earning bachelor’s degree recipients out of graduate school, it would tend to depress estimated returns to graduate work. If true, then least squares estimates of the returns to graduate school that ignored the role of major-specific ability measures would tend to understate the true returns. Our empirical work provides evidence consistent with this sorting story.</blockquote>

    <blockquote>GRE scores have an interesting impact on the probability of pursuing a higher degree. Undergraduates in majors with higher verbal scores and lower quantitative scores are more likely to pursue the doctorate or professional degrees. The standard deviation of GRE scores in the major tend to reinforce the effects of the mean scores: higher standard deviation of GRE verbal scores raises the likelihood of pursuing the doctorate, while increasing the standard deviation of the quantitative score lowers the likelihood of pursuing the doctorate.

    <blockquote>As shown in Figure 3, the proportion of students stopping at the bachelor’s degree rises and falls with the average quantitative GRE score, while the likelihood of seeking doctoral or professional degrees is negatively related to the quantitative GRE. The result from Table 3 that the marginal impact does not vary across graduation cohorts suggests that this is a result of rising quantitative skills and not rising returns to those skills.</blockquote>

    <blockquote>Assuming a Master’s program takes two years and a PhD program takes 6, implied annual returns are only 5.9% and 4.3% respectively.10 Annualized returns to professional degrees are more reasonable at 15.1%, assuming a four year program. There is a significant positive return to GRE mathematics scores, but no measurable return to verbal skills. There is a significant premium for postgraduate degrees in business and a significant discount for postgraduate degrees in the sciences.</blockquote>

    <blockquote>Those who do not go on to graduate school are drawn atypically from the upper tail of the GRE quantitative distribution and the lower tail of the GRE verbal distribution, both of which are expected to raise their earnings. On the other hand, those who go on to graduate school are drawn disproportionately from the lower tail of the quantitative GRE distribution and from the upper tail of the GRE verbal distribution, both of which lower their opportunity costs of graduate school. Consequently, the observed premium of average earnings for post graduate degree holders over bachelor’s degree recipients understates the true returns to graduate school. Correcting for the sorting raises the estimated returns, as found in Table 4. </blockquote>

    <blockquote>This leads to a sorting effect whereby students whose cognitive skills would suggest lower earnings at the bachelor’s level are more likely to attend graduate school. This sorting effect appears to be part of the cause of the downward bias in estimated returns to graduate education—the average earnings of those who do not go to graduate school overstate the opportunity costs of graduate education for those who do pursue advanced degrees.</blockquote>

  8. floyd says:

     The headline should read “Americans Under-Degreed”, considering the content of the article. 

  9. George Leef says:

    It’s misleading to look at average earnings statistics, which include many people in the college category who earned their degrees decades ago and are at their income peaks now. The fact that on average those who have earned college degrees make a lot more money does not mean that young people earning college degrees today will enjoy the same results. This argument is usually given to support the notion that the country needs to put more people through college, but we’re already scraping the bottom of the barrel with regard to academic interest and ability. Many high school grads enter college with miserable basic skill levels and even if they graduate, they’re apt to wind up in rather low-skill, low-pay jobs. College adds to the human capital of some students, but for many others it adds almost nothing except for debt and merely delays the point in time when they enter the labor force and begin to learn something useful.
    The reason why the earnings gap has continued to rise is not because college graduates are so much smarter and more capable than those who only complete high school, but rather because of credential inflation. That is, many employers now require that applicants have college credentials for jobs that only call for basic trainability. As a result, good career paths for non-college people have been shrinking. That is why average earnings for workers without college degrees has been pretty stagnant.
    College is terribly oversold. Only a very wealthy nation could afford to have a higher education system that costs as much and delivers as little as ours.

  10. just me says:

    Honestly, I would be afraid to take on the kind of debt required to get a degree right now if the job prospects after graduation aren’t looking so good.
    My oldest will be going to college next school year and a huge part of choosing a school will come down to which school gives her the most money to attend rather than which school she most wants to go to, because of the debt involved.  And she is smart enough that she can expect some merit aid.  I can’t imagine what somebody who graduates somewhere in the middle with mid to low test scores is going to do-especially if they don’t fall within the right incomes for need based aid.
    That said, i am not convinced everyone needs a four year degree.  I think there are a lot of jobs that a two year community college or tech school can train for at a lower cost than college with a better rate of return.  A plumber in our area, once fully certified can make about 60,000 to 80,000 dollars a year, because there aren’t very many of them and it is a skill that many people don’t have.
    Going to school for training as a plumber and meeting the various license requirements in the end has a better income return for less student debt than a four year degree does.  And around here at least a plumber doesn’t have to worry too much about finding work.