The Bureau of Labor Statistics has released new employment projections for 2022. These projections are frequently misinterpreted, and the way BLS presents the data can certainly leave the uninitiated confounded. This analysis uses the occupation projections to discuss two issues: whether low-, middle- or high-wage occupations will grow disproportionately, and whether the occupation structure of 2022 relative to 2012 requires substantially more education and training. The answer to both questions is that the occupational structure of 2022 does not look dramatically different than what we have now. This means that the challenge we face is how to make occupations better paid rather than worry about whether the workforce we have is under-skilled or over-skilled for future work. That is, we have a job quality problem, and not a skills deficit problem. (Becky Thiess reached the same conclusions in an analysis of the prior set of projections.)
These projections get used in misleading ways. Some people look at the occupations that grow the fastest and draw conclusions, usually that we all need a lot more education. Others emphasize which occupations create the most number of jobs and find that a large expansion of low-wage work is looming. The BLS press release ricochets back and forth between both approaches, which obscures what one should conclude from the projections. Neither approach—looking solely at either the rate of employment growth or the absolute amount of employment growth—is correct. A fast growing occupation may be relatively small, and therefore inconsequential in the overall economy. A large occupation may generate a lot of absolute employment growth, but if it grew at the average of total employment growth (so its share of total employment did not change), then the overall character of employment (more skill, higher/lower wage) would not change.
The way to cut through this noise is to examine which occupations grow and shrink as a share of total employment: it is the difference in the character of occupations with increasing or decreasing shares of employment that establishes how the future differs from the past. An occupation that grows from 6.0 to 6.2 percent of employment matters as much as one growing from 1.0 to 1.2 percent of employment (and the former grew more slowly than the latter). Anywayyyyyyyy, that’s why we look at changes in occupation employment shares.
We first look at changes in occupation employment shares by the wage levels of occupations to see if the projections reflect the familiar job polarization story (see Don’t Blame the Robots for a critical examination of this narrative of the economy). That is, are low-wage and high-wage occupations expanding while middle-wage occupations are shrinking? The table and graph below show the share of employment in 2012 and 2022 in occupations grouped by wage level: we first defined the lowest-paying occupation group as the bottom twenty percent of employment ranked by wages; the lower-middle occupation group comprises the rest of the bottom half; high-wage occupations are the top-paying ten percent with the upper-middle group comprising the remaining part of the upper half. No big changes in the occupational structure are evident here, as each wage group’s share of employment remains pretty much the same in 2022 as it was in 2012. We are not faced with a future of employment in low-wage occupations (20.02 percent of employment in 2012, 19.99 percent of employment in 2022) nor is there a sizeable employment increase in high wage occupations (high wage occupations grow from 10.02 percent in 2012 to 10.28 percent in 2022) that people can shift into.
Occupation distribution by wage level, 2012- 2022
Note: occupations are ranked by median annual wages in 2012. Occupation distribution excludes actors, dancers, musicians and entertainers due to the lack of information on wages.
Source: EPI analysis of Bureau of Labor Statistics, Employment Projections (2013)
BLS does a nice job of presenting how occupation employment shifts affect the need for various levels of education. We’ll just focus on the implications for what share of employment requires a college degree or further education. BLS shows that the number of jobs in occupations that required a bachelor’s, master’s or doctoral degree comprised 22.3 percent of all jobs in 2012, and are projected to rise to 22.8 percent in 2022. That’s not much change or challenge. For one, roughly a third of the workforce already has a college degree or further education. Secondly, the share of the workforce with a college degree or further education has grown steadily for decades, rising, for instance, by 5.7 percentage points between 2000 and 2011. Obviously then, an increase of 0.5 percentage points over the next ten years is not much of a challenge (especially when, taken literally, the BLS data suggest we already have too many college graduates).
Our conclusion is that the shifts in the type of work expected in the future, as indicated by occupational employment projections, are far from transformational or are even much of a change at all. The types of jobs we will have in ten years are pretty much the types of jobs we have now. Whether workers are able to earn more or obtain better benefits will depend on how the job quality of particular occupations changes over the next ten years, and will not depend on which occupations are growing in importance.