Unemployment is at an all-time low! Great news, right? Not so fast. GETTY
You might have noticed, in the last several months, that a curious trend has begun to take place in economic news. Unemployment has hit a fifty-year low - the last time it was this low, 3.7%, was in 1969, when the big news was not the economy but the Vietnam War. Yet if you are paying attention to the job payroll numbers, you may also have noticed that the number of new jobs created has been dropping slowly but steadily to 136,000 in September (compared to the average of about 250,000), with both July and August numbers also revised downward.
This would seem to be a paradox - both of these shouldn’t be true at the same time. In actually, it is quite possible for both to be true, because they measure different things, but the spread is also an example of hyper-employment, which is not as benign as it may appear on the surface.
The US Bureau of Labor and Statistics (BLS) has six different metrics for measuring employment, commonly referred to as U1 to U6. U1 is the broadest measure and indicates the number of people who were unemployed for fifteen weeks or longer divided by the total labor force, where unemployed here means having no wage income during the period in question. U3 is often referred to as the headline rate, as this is the rate that is most frequently quoted in the news. It’s the unemployment of people who are not currently working but are seeking work, and that have been unemployed for at least four weeks.
When you dig into these definitions further, what becomes fairly obvious is that they are remarkably ambiguous. If you work in an office for forty hours a week (give or take) then it’s easy to see that you are employed. However, what happens when you are working only ten hours a week? You’re still employed by the u3 definition, even if you’re not making enough to live on. Suppose that your income comes from royalties (such as that of a writer or podcaster)? That gets fuzzy. Suppose that you live off a trust fund? Technically, you’re unemployed. Sell drugs? Unemployed. The problem with boiling down a concept such as unemployment in a single number is that it usually hides a great deal of ambiguity.
Note that measuring employment is remarkably difficult and is becoming more so as the definition of work in the gig economy challenges many existing models. GETTY
What this means in practice is that most measures of U3 are based upon statistical surveys, both of individuals and companies. Did a person work at least X number of hours in the last month? How many people did a company hire and fire this month? How many applications for unemployment were made (and granted)? Periodically, the BLS may change their survey methodology, at which point, they should publish warnings indicating that data gathered from this month to the next may prove to be wonky.
As a consequence, while politicians may crow about low U3 measures, most economists tend to treat them with a certain degree of caution, simply because the definitions are so vague.
Payroll numbers are a bit different, though again these need to be taken with a grain of salt. Payroll numbers typically can be determined by monthly filings, and are in effect the number of people who gained employment at a company minus the number of people who lost employment at that company. U3 numbers are driven by intent - is someone looking for a job or not. Payroll numbers, on the other hand, are more definitive; though usually don't indicate the reasons why people left the company, such as better opportunities elsewhere, retirement, layoffs, firings for cause and so forth. Additionally, payroll measures are still statistical in nature and tend to favor stable, mature companies over startups. Gig economy jobs are almost invisible here.
One additional factor in interpreting labor statistics is that inflection points (near the top and the bottom of a market) usually tend to give off many false positives. At the top of a market, job hopping becomes frothier, as the quality of jobs becomes marginal and the length of those jobs decreases. Contract jobs that had been for a year or more become six month, then four, then two. More and more work is moved through contingency agencies, as companies, worried about headcount and margins prefer to keep contracts short and managed.
Economists refer to this as churn, and at the moment churn is high. This can be seen by the fact that while wages are finally rising after having been largely static for the last decade, they aren’t rising very fast. Churn often gives the illusion that there’s a lot of economic activity taking place because of the amount of hiring, but the lack of wage pressure strongly suggests that many moves are lateral, with no corresponding rise in wages from one job to the next. During the last equity led recession (2001-03), churn played a big part in creating a similar hyper-employment scenario.
Churn is a measure of the turbulence in the job market. The higher the churn, the poorer the quality of jobs, and the more likely that a market top is near. KURT CAGLE
Full employment is typically defined as being when U3 is at 4.0%. This does not mean that 96% of the workforce is fully employed; only that 96% are receiving some form of wage income. If measured in terms of Full Time Equivalent (FTE), there are about sixty five full-time jobs for every hundred people in the workplace, with forty-five to fifty of those being 37.5 hours a week or more (the definition of full time). Sometimes part-time jobs are held voluntarily - a mother working while her kids are in school, a student working around classes, a retiree supplementing their pension, an actor working as a waiter between performances and so forth. More often, that may be all that a person can get, which is typically true later in an economic cycle where the quality of jobs diminishes.
The size of the active labor force also changes. In 2018, half of all baby boomers turned sixty-five. This population spike is now manifesting in the reduction of the labor force and will continue to do so for the next fifteen years or so. Not every boomer is retiring, of course, but a large enough percentage is that the labor participation pool will continue to shrink, meaning that employment, relative to the size of the labor pool, will stay surprisingly high.
This is one of the reasons for the seeming hyperemployment paradox. The size of the labor pool has a direct impact upon the amount of economic activity that takes place - the larger the pool, the more people there are to do work (especially in a knowledge economy) and the faster the economy grows. A shrinking labor pool likely indicates a shrinking economy; regardless of current economic policies are in play.
The velocity of money also slows down in that same scenario. When the Federal Reserve injects a dollar into the economy, that dollar will be spent multiple times, or put another way, each "created" dollar translates into ten, twenty, fifty or more "used" dollars. The smaller the labor pool, the fewer the number of interactions that this dollar gets used for, and the weaker the stimulus. With poor monetary policy, this can, in turn, lead to inflation, as there are so many dollars in circulation that the value of those dollars drops. As prices rise relative to wages, economic activity in the labor pool also slows down, leading into a self-reinforcing spiral.
The velocity of money is a direct indication of economic activity. The faster that money passes from one hand to the next, the greater the economic activity.
The investment class runs into a similar problem. In absolute terms, the investment class may spend far more on a per-capita basis, but relative to their numbers, once a dollar is captured at the upper echelon it tends to stay in the upper echelon and ends up taking working capital out of the economy. (It's the primary reason why taxes on capital gains are quite effective at stimulating the economy, even if they are hated by the investment class). This is the fundamental fallacy of trickle-down economics - the gradients for money flow moving upward far exceeds the gradients that come from an investment that moves downward. Wealth concentrates faster than it diffuses.
The number of job types in the manufacturing sector is declining even as the sector itself is shrinking relative to the economy. This is happening globally; forty years of innovation in manufacturing have reduced the required footprint of people necessary to build things, and this will continue for the near future. The professional services field (what I’d peg as being early knowledge workers) is also shrinking: you need fewer account managers, fewer lawyers, fewer accountants, fewer professors. The collapse of Thomas Cook recently highlights this - travel agents have been struggling to redefine themselves since the early 1990s, but the field is collapsing faster than new jobs can be created.
At the beginning of the Internet, in the 80s and early 90s, traditional jobs disappeared in greater numbers than they have today, but augmented jobs, harnessing computers to create new job categories (e.g., Experience Professionals!) caused that economy to boom and kept the overall rate of attrition relatively low.
However, we’re now moving to a stage where augmentation is giving way to replacement. What employers are looking for are highly skilled specialists, but to create a specialist, you generally need an infrastructure to train that specialist, a pool of trainees capable of learning those skills, and time for them to master that domain. They’re also looking for cheap, which has generally meant offshoring.
What is happening now is that immigration (including technical work visa applications) is drying up. This is already creating a situation where the cost of technical talent in the US is jumping dramatically as the availability of candidates drops, which is, in turn, creating a shock wave in the labor market. The result is the hyper-employment scenario we're facing now, as companies can no longer find the candidates necessary to keep them competitive.
I expect that there may be a short purge in the knowledge space (it is somewhat overbuilt at the moment) but that this will likely have comparatively little impact upon knowledge workers. Outside of this field, however, we may be looking at what happened to companies like Thomas Cooke and Sears as being the new normal for a while, as the augmentation to replacement wave continues.
Wage growth and hours worked may provide a better indicator of the health of the economy than measures such as GDP or unemployment numbers. GETTY
In the interim, watch wage growth and the number of hours worked. If they grow significantly, then we may have dodged a bullet. However, if they stall, or worse, decline, then it is very likely that the economy is going to go south quickly.
Follow me on LinkedIn.
Kurt Cagle is Managing Editor for Cognitive World, and is a contributing writer for Forbes, focusing on future technologies, science, enterprise data management, and technology ethics. He also runs his own consulting company, Semantical LLC, specializing on Smart Data, and is the author off more than twenty books on web technologies, search and data. He lives in Issaquah, WA with his wife, Cognitive World Editor Anne Cagle, daughters and cat (Bright Eyes).