There’s a mystery at the heart of the U.S. labor market.

Don’t fret over the -33,000 month-to-month total jobs change in September. There was a unique circumstance. Work in southern Texas and much of Florida was idled as Hurricanes Harvey and Irma paid unwelcomed visits.

The jobs number will bounce back aggressively in October. (It should be +180,000 at least.)

The U.S. unemployment rate at 4.2% is as low as it has been in 16 years, dating back to the early 00s.

The weekly initial jobless claims figure for the week ending October 14th was only 222,000, its best reading in 45 years, dating back to the early 1970s.

And yet, employee compensation has failed to take flight?

According to the Bureau of Labor Statistics (BLS), year-over-year average hourly and weekly earnings for all workers in the U.S. economy have been inching up, but they remain no greater than +3.0%. Construction workers have been doing a little better, but not by much.

It has been reported that Janet Yellen, Chairman of the Federal Reserve, is a huge fan of the Job Openings and Labor Turnover Survey (JOLTS report) as a source of insight into what is transpiring in the U.S. labor market.

Therefore, a close examination of JOLTS results for job ‘openings’, ‘hires’ and ‘quits’ is warranted. The six accompanying graphs, for all workers first and then for construction sector employees specifically, will aid in the investigation.

Graph 1 shows that total job openings, expressed as both a level and a rate, are at extraordinarily high levels. They’re well above their 2007 pre-recession summits.

In Graph 2, featuring job hires, the curve for ‘level’ has returned to a quite advantageous spot. The curve for ‘rate’, however, while hovering in a healthy range between 3.5% and 3.8% over the past three years, has not attained the heights it achieved in 2004-2006. Also, it is well short of its early 2001 performance.

Graph 3 highlights ‘quits’ data. Understanding the relevance of ‘quits’ is counter-intuitive. The emphasis isn’t on the degree of annoyance employees feel concerning their current situations. Rather, the focus is on how confident they are that if they leave their present positions, they will be able to find something that is a step up.

According to the latter interpretation, a greater incidence of ‘quits’ signals improving labor market conditions that are generating better alternative opportunities.

The total-jobs ‘quits’ level in Graph 3 is elevated and continuing to climb. The ‘quits’ rate is also elevated, but it has leveled off over the past year, at around 2.1%.

Since the lack of action in the ‘quits’ rate can be explained by the plateauing of the ‘hires’ rate, the relevant question is why the ‘hires’ rate has been struggling to make headway?

One explanation may be found in the supply of labor. The labor force participation rate among all individuals aged 16 and over has fallen from 67% in 2000 to 63% at present.

This suggests there are too few people to fill the vacant positions. Furthermore, the shortcoming is further exacerbated by a skills gap. Many current openings require skill sets (e.g., in high-tech) that a disproportionate number of today’s out-of-work individuals do not have.

There is no increase in wages that can solve a labor availability issue based on a skills shortfall.

Next, it is the turn of construction to be examined from a JOLTS perspective.

From Graph 4, it is apparent that U.S. construction job openings measured as both a level and a rate have picked up significantly. They are now back to where they were in 2007 when the economy was submarining through quiet waters with nary a worrisome blip (i.e., having to do with the coming recession) on sonar screens.

The JOLTS ‘hires’ numbers for construction (Graph 5), however, tell a different story.

When placed in a historical context, neither the current ‘level’ nor the ‘rate’ of construction hires is particularly remarkable. The ‘level’ has barely budged in almost nine years. It was considerably more noteworthy from 2001 through 2006.

The ‘rate’ of construction hires since 2013 has also underperformed and has not risen to meet the standards established from 2010-2012, nor from 2001-2005.

Turning to Graph 6, the quits ‘level’ and ‘rate’ in construction have been on gradually ascending inclines. But again, they have not yet attained their prior peaks in early 2006 and early 2001.

While there has been much talk of contractors bidding up the price of labor, the collateral impacts on the JOLTS ‘hires’ and ‘quits’ numbers have, so far, been limited.

One key segment has kept the construction sector from having more glowing JOLTS results.

U.S. homebuilding has been failing to live up to its potential. Based on population growth and historical rates of family formations, there has been a persistent shortfall in housing starts for a decade.

Disappointing residential foundation work eases some of the pressure on what would otherwise be an extremely tight construction labor market.

Other aspects of the demand-supply equation for construction labor, however, have turned problematic.

The construction sector’s traditional answer to meeting out-of-the-ordinary labor needs has been immigration, legal or otherwise. For reasons having to do with the new Washington administration’s policies, this is no longer as clear an option.

Also, there is an uncertain but immensely impactful socio-economic topic to be explored.

Much hope for the future of the U.S. economy is being placed on the millennial generation. A great many millennials are on the cusp of leaving cloistered lives (e.g., in academia while living in dorms or with their parents/grandparents) to enter the workforce.

The participation rate of millennials, after ten years of declines, has begun to move higher again. During the past year-and-a-half, it has shifted from 81% to 83%.

Even if this does eventually help to alleviate the labor shortage problem for the economy as a whole, it will leave the construction sector with a dilemma – how to convince millennials to choose fieldwork as a preferred career path.

When asked in ‘intentions’ surveys, millennials say they are just as motivated to buy single-family housing as their parents were. If that turns out to be true, the kick-start to homebuilding will propel national output growth for many years to come.

But here’s a final thought. Many millennials starting out in the workforce are opting for temporary gigs as opposed to making career commitments − hence the recently coined term the gig economy.

Will earnings from gigs pay the mortgage?