Comparing a Decade of Income Gains by Education Level

We all know that, as a general rule, the more education an occupation requires, the better it pays. It is also commonly understood that the advantage of a college degree has been growing. This has been demonstrated by surveys that ask people—especially recent graduates—about their education and about their earnings. But how do income gains compare across different levels of education when you look at the issue not through surveys of individuals but rather through groups of occupations?

To answer this question, I looked at the income medians reported for occupations by the Occupational Employment Statistics (OES) program. I grouped the occupations by level of education required as specified by the Department of Labor. Then I computed a weighted average for each level; in a weighted average, the occupations with larger workforce size count for more in the average. I did this for two years: 2002 and 2012.

Here’s what I found:

Education Level	Weighted Average for 2002	Weighted Average for 2012	Gain
High School	$25,119	$37,799	50%
Vocational Training	$29,565	$36,530	24%
Associate Degree	$42,942	$56,917	33%
Bachelor's Degree	$57,137	$71,756	26%
Master's Degree	$44,155	$65,532	48%
Doctoral/Professional Degree	$ 81,022	$106,915	32%

I’ll admit that I was surprised by some of these results.

One surprise was the low payoff for master’s-level occupations—a weighted average of $65,532 in 2012, which is lower than the average for bachelor’s-level occupations. Understand that the set of master’s-level occupations includes many workers who are low-paid considering their required degree. For example, out of 2,197,830 workers, it includes 104,070 Rehabilitation Counselors, with a median income of $33,880; 115,080 Mental Health Counselors ($40,080); and 99,680 Community and Social Service Specialists, All Other ($41,060).

These average figures (both for 2002 and 2012) for master’s-level occupations defy the general trend that the more education you have, the better you’re paid. These figures also contradict the chart created by the Department of Labor showing median weekly earnings of individuals at various education levels, in which people with a bachelor’s average $1,108 per week and those with a master’s average $1,329. (The figures apply to 2013, but you would not expect much difference from 2012 data.) I think the reason might be because the Rehabilitation Counselors and other master’s-level workers with comparatively low annual earnings are more likely to work part-time, and therefore their advantage in weekly earnings would not be reflected in their annual earnings.

Something similar is probably at work in the ho-hum gain (32 percent) for occupations that require a doctoral or professional degree. A large number of these workers are college faculty, and the current trend on campuses is toward greater and greater use of adjunct—that is, part-time—instructors.

The other surprise was in the percentage gains in income. We’ve all heard that wages have remained flat for workers with low education levels, yet the weighted average of high school–level occupations show the biggest gains between 2002 and 2012, and the bachelor’s-level occupations show the second-lowest gains. I think that the reason we find bigger gains for the bachelor’s degree in studies like that done by the College Board is that those studies look at comparatively recent recipients of the degree, whereas the figures I used are based on all incumbents in the occupation. The set of all incumbents includes many workers who have been in the occupation for 20 or 30 years; some of these may have more or less education than the degree that is now recommended for the occupation.

Therefore, young people planning their education should not base their decision on my analysis here. What I show here are trends based on sets of workers that include only small numbers of recent graduates.

Another consideration to bear in mind is that my figures apply only to workers who are employed. They do not reflect the large losses of income experienced by workers in high school–level occupations who became unemployed, as so many did, especially as a result of the recent Great Recession. The large gain 2002–2012 for the high school–level occupations may in fact reflect a kind of Darwinian selection: Workers in these occupations who remained employed despite the Great Recession are probably the ones with the highest levels of skill.

Job-Growth Projections for Men and Women

Today the Senate failed to pass legislation aimed at closing the pay gap between the sexes, and a lot of attention is being paid to the extent of that gap and what can be done about it. I have blogged about this issue several times, but for this week I decided to look at a different male/female issue: Who has the better outlook for job growth?

To answer this question, I assembled figures from the Bureau of Labor Statistics on the percentage of men and women in various occupations and the projected workforce growth (2012 to 2022) of those same occupations. Then I calculated the correlation between male/female presence and job growth.

Before I tell you my findings, here are two important caveats: (1) Figures for percentage of women are not available for many occupations. In most cases where figures are lacking, it is because the occupation is so heavily male that the sample of women incumbents is statistically insignificant. So instead of basing my correlations on detailed occupations, I based them on 22 families of occupations—for example, Management Occupations and Food Preparation and Serving Related Occupations. (2) My correlations are based on the present representation of the sexes in the occupations, not what the percentages will be by the time the projection period is over. It is possible that by 2022 we may see different gender ratios in some occupations.

Okay, now for my findings: The correlations between these occupational families and their outlook worked out to be 0.47 for women and -0.47 for men. In other words, there is a significant tendency for female presence in an occupational family to predict expansion of the workforce, whereas male presence tends to be linked to a shrinking workforce.

If you have been following trends in the economy, these findings should not be surprising. The fastest-growing (28.1 percent) segment of the economy is Health-Care Support Occupations, and many of the fastest-growing occupations in that segment are those (such as Registered Nurses) that are heavily dominated by women. Women are also prominent in another fast-growing (20.9 percent) segment, Personal Care and Service Occupations.

If you delve down to a level of greater detail, the correlations seem to be much less strong. I ran correlations using detailed occupations for which female percentages are reported and found results that were vanishingly close to zero. Now, understand that to create these calculations, I had to throw out a large number of occupations that had no female percentage reported, and in many cases these discarded occupations were virtually all-male. Many of them were in the Production Occupations sector, so I lost much of the drag of this sector (0.8 percent growth) on male prospects. On the other hand, many of these virtually-all-male occupations were in the Construction and Extraction Occupations sector, which is second only to Health-Care Support Occupations in projected growth (21.4 percent).

All things considered, I believe it likely that at a more detailed level, the relationship between gender presence and outlook is more tenuous. So if you seek an occupation with a good outlook, make a point of learning the growth projected for that specific occupation, and don't base your expectations solely on the female or male presence in the occupation.

Employment Trends for Engineers

As the “E” in “STEM,” engineering is an important field of knowledge that is vital to the American economy. I thought it would be interesting to see how employment in this field was affected during recent years of recession and recovery, so I created the following graph (from BLS figures):

My main takeaway from this chart is that employment in engineering occupations tends to be reasonably stable. Apart from the 2007–2008 employment decline of 26 percent for aerospace engineers, only one occupation saw a double-digit decline between any two years: the 12 percent 2008–2009 decline for computer hardware engineers. On the other hand, these occupations did not show great growth over this period either—with the exception of petroleum engineers, which grew by 127 percent. Their average growth over the 2007–2012 period is only 3 percent, if you remove petroleum engineers.

Although employment in the occupations changed within relatively modest boundaries, the fluctuations show quite distinct paths. Employment for most of them bottomed out in 2010, evidently the delayed result of the recession as employers ran out of rainy-day funds and needed to lay off workers.

Here’s another chart, with the same set of occupations, showing projected employment growth between 2012 and 2022:

Petroleum engineers is the standout occupation here, again. The recent upward slope of civil engineers resembles that of many of the other specializations, but the growth of this occupation really takes off in the decade ahead. The Occupational Outlook Handbook reports, “As infrastructure continues to age, civil engineers will be needed to manage projects to rebuild bridges, repair roads, and upgrade levees and dams. Moreover, a growing population means that new water systems will be required while the aging, existing water systems must be maintained to reduce or eliminate leaks of drinkable water.” By comparison, the projected growth for mechanical and industrial engineers is modest.

Over the course of a career, engineers tend to find advancement by going into management (for example, project management) rather than by becoming increasingly specialized. Employers often are reluctant to increase the pay of seasoned engineers when they can hire young graduates of engineering schools who are conversant with the latest technologies. Sales engineering is another potential route for career development, especially for those who have good people skills and are not averse to frequent travel. Technical writing is another possibility, but it is much less lucrative.