A recent article at CEPR estimates the change from 1979 to 2011 in the proportion of Low Wage Earners at different levels of education (and age, but my focus is only on educational attainment).  I can't find their 1979 source data at the gov't source they cite, but I do have the employment by wage & educational levels for 2000 and 2011 for Full Time employees over 25 (and the intervening years) from the same gov't source.  I assume CEPR's data includes all employees ... full and part-time, age 16 and over, so CEPR's data would tend to show a greater proportion of low wage earners than the data for full time & over 25 workers.  Both are based on 2011 real dollars.

The data source I used shows wage/salery for: 1st Quartile, 2nd Quartile (Median), 3rd Quartile, and upper decile.   With that income data I can only approximate the Low Wage percentile where Low Wage is equal or less than $10/hr (CEPR's definintion). 

The upshot is that my estimates turn out to be close enough to CEPR's that I can reasonably assume my estimates for 2000 are also valid.  ... taking into consideration that I'm using Full time Employment over Age 25 while CEPR's numbers assumedly include all employees (including part time & age 16 & up). 

In any event, the data by CEPR and my estimates from 2000 tell the same story.... a greater proportion of more highly educated employees are working for Low Wage (=< $10/hr) in 2011 than they were in 2000 (or 1979)  

The data shows that there are now 12% more full time employees in the "Low Wage" bracket than there were in 2000. Of the Low Wage full time earners, there are now 1.3% fewer with < High School education, 5.6% more of High School grads (no college), 46% more with High School + some college, and 4% more with a 4 year college degree or more.  

At 2011, of those in the Low Wage bracket, 25% have < High School education, 38% have High School degree (no college), 28% have High School + Some College, & 9% are 4 year College grads or better. 

The "Low Wage" percentile cuttoff in 2011 for each educational level is (2000's cuttoff in parenthesis):

38% < High School (31%)

16% High School only (13%)

12% High School  + Some College (8%)

 6%  4 Year Bachelors and better (7%)

In other words, of all workers with less than a High School diploma 25 years or older, and who are full time employees , 38% earn $10 or less/hr.... etc.    The reall problem though is that there are more now than in 2000 for each educational attainment level (about the same for 4 Year Bachelors and better).  This stands to reason as there are a limited number of high wage/salery jobs for higher levels of education in proportion to the number with higher levels of education now more than was the case in 2000 (and in 1979 also for that matter).  This means more of the better educated have to take the more widely available lower wage/salery jobs.

Since this condition's progression has been going on since at least 1979, with the data in 2000 showing it still occurring relative to 2011, then the phenomena is on-going.  What this means is that the greater the proportion of higher levels of educational attainment achieved the fewer available slots there are for employment commensurate with the level of education.... growth in demand for employees with higher levels of education isn't keeping pace with the growth in supply... supply exceeds demand... hence this keeps a lid on wage and salery growth --- as my prior note showed, real wage/salery incomes since 2000 have declined except for those with advanced degrees (also see table at end of this note).     

Here are my numbers:

Table of Changes in wage / salery by educational attainment since 2000: