Olly, olly oxen free
admin | February 14, 2020
This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors.
Shout “Olly, olly, oxen free!” on most playgrounds, and all the kids playing hide and seek can step out into the open. Something like that happened in the labor market in 2019. People classified as out of the workforce came out of their hiding spots to accept positions that firms had struggled to fill. And now the January employment report suggests more of the same. The key question going forward is how many more potential workers are hiding in the shadows and waiting for a new call into the workforce. Both growth and inflation hinge on the answer.
Labor Force Participation
The US labor force participation rate is clearly on a long-run downward path. Chalk it up to an aging population as well as Labor Department calculations. The department defines the working-age population as 16 and over, meaning there is no age cap. As the massive cohorts of Baby Boomers retire, the aggregate labor force participation rate consequently will move lower. But a substantial cyclical inflow of workers during a period of robust labor demand has temporarily arrested that downtrend.
One way to control for the changing age composition of the population is to create a theoretical age-constant labor force participation rate. The Bureau of Labor Statistics publishes breakdowns by age for all of the key labor market data from the household survey, including working age population, labor force participation, and employment. It is a relatively simple exercise to pull the BLS figures for the labor force participation rates for each age segment. Rather than apply the actual, evolving weights of those age segments, instead use a constant set of weights. This measures the aggregate labor force participation rate as if the age composition of the population stayed constant.
Using this approach, the age-constant LFPR has surged over the past two years (Exhibit 1). In fact, the January 2020 reading of 66.8% matches the best level since 2001 and is very near the all-time high of 67.2% registered in 2000.
Exhibit 1: Labor force participation: actual and age-constant
If the all-time high reached in 2000 represents a hard ceiling, then the remaining margin of people available to come off the sidelines and into the labor force is roughly one million, a small safety valve compared to the 3.25 million that came off the sidelines over the last 25 months.
Not in the Labor Force but Want a Job
Within the monthly household survey, the source of all these measures, respondents answer a series of questions to classify them. For those who say that they are not working and did not actively look for a job in the past month, classifying them as out of the labor force, a follow-up question is whether they would in theory like to work. One can imagine a number of reasons why someone might be amenable to working but was unable or unwilling to actively look for a job in a given month: health issues, family obligations, transportation issues, weather, or a perceived lack of opportunities.
Those who indicate that they would in fact be interested in a job but did not look over the past month and thus are not considered unemployed are reported within the household survey figures. To be clear, this grouping is clearly a subset of the universe of people who can come off of the sidelines to work, but the series offers a second way of thinking about how much margin remains to pull new people into the workforce. The series has fallen in recent years but is not quite down to the lows seen in the late-1990s (Exhibit 2) To be fair, the labor force and working-age population are also substantially larger today than they were 20 years ago.
Exhibit 2: Pool of workers not in the labor force but interested in working
Pool of available workers
Combining those classified as unemployed and those who were outside the labor force but want to work offers an estimate of the pool of readily available workers. We can then create a modified unemployment rate by dividing that pool by the sum of the labor force and those outside of the labor force who are amenable to working. That modified unemployment rate fell to as low as 6.8% in the late-1990s and early 2000s, when the unemployment rate bottomed out at 3.8%. In the current expansion, this measure reached the prior all-time low of 6.8% in the second half of 2018 and has continued to decline rapidly, sitting in the 6.2%-to-6.4% range over the past five months.
Exhibit 3: Broader unemployment rate
If labor demand remains robust in 2020, the key question will be whether firms are able to continue to pull large numbers of job candidates off of the sidelines or whether the economy begins to run short on people. In the former scenario, the unemployment rate could still inch lower, as it did in 2019, but job growth can remain solid and wage gains could stay moderate.
Alternatively, if the flow of people into the labor force begins to dry up, then job gains will necessarily slow and GDP growth will probably cool as well. At the same time, presumably, in that environment, firms would become more aggressive at bidding up job candidates, driving wage growth significantly higher and perhaps pushing prices up as well.
The Fed’s willingness to push the envelope, allowing the unemployment rate and labor market slack more generally to continue to fall until alarm bells actually go off, probably means that at some point, we will in fact start to hit the proverbial wall. The labor market got awfully tight in the late 1990s, and, as the data explored in this piece suggest, we are currently close to that reality today, perhaps with still a little more slack to play with depending on which variable you choose to focus on. I am skeptical that firms will have as much success in 2020 finding new people to hire as they did in 2019.