In my recent piece on U-6, I mentioned that the labor force participation rate (LFPR) is a favorite target for critics of the economic recovery (and the Obama administration). I mentioned that I’d written about the LFPR quite some time ago, and that Bill McBride had done likewise at Calculated Risk:
What is probably my most comprehensive piece on the LFPR is here, published in May 2014. Bill McBride, over at Calculated Risk, has also done some work on this file (see here and here). (Bill is one of the best economic bloggers out there, and I have a ton of respect for his work and am thrilled to have gotten to know him a bit over the years.) LFPR out of the way, let’s move on.
Bill has referred to critiquing the LFPR as “the last refuge of scoundrels.” And with good reason. I had no intention of revisiting LFPR, but a recent piece of research changed my mind, as you’ll soon see.
In what I believe is her most recent piece (December 2013) on the topic, BLS economist Mitra Toossi – probably the foremost authority on the subject – projects that the LFPR will continue to decline, incorporating this chart into her work:
But enough about Bill McBride, Mitra Toossi, and me. My good pal Dave Rosenberg weighed in on the LFPR in the November 12 edition of his always fabulous Breakfast With Dave. If you subscribe to one top-notch, top-down, daily view of the world, you should make it Dave’s – it’s easily the best daily insight out there, and you can scrap five others in favor of his.
Oh, and those of us who know Rosie personally found Ritholtz’s interview with him (yesterday, here) to be both informative, with parts that are hilarious.
This is a longish, thorough, and informative read, so sit back and enjoy. And remember, no credible talking head would cite this stat. Keep this in mind as you watch various business channels and shows. Continue on, then impress friends and family at upcoming Thanksgiving gatherings with your newfound knowledge!
Reproduced here with the author’s permission. Take it away, Rosie:
They ain’t coming back
There has been a surprisingly large amount of commentary writing off the improvement in the U.S. unemployment rate — which dipped below the Congressional Budget Office’s estimate of the non-accelerating inflation rate of unemployment (NAIRU) for the first time since February 2008 in October — given that we have not seen an attendant uptick in the labour force participation rate.
The thinking behind this makes some sense on the surface — the traditional unemployment rate focuses on the labour force, so in a situation where there is a huge whack of people falling out of the labour force due to discouragement over job prospects (i.e. not counted as “unemployed” and not included in the labour force), we would see a decline in the unemployment rate but no improvement in the participation rate.
And this line of thinking is consistent with what we are seeing in the U.S. right now with the unemployment rate touching cycle-lows at 5.0% in October while the participation rate sits at a 40-year low of 62.4%.
But, as we have talked about repeatedly before, the fact of the matter is that those that refer to themselves as “discouraged workers” represent only a very small fraction of the overall population (just 665,000 people or a tiny 0.26% of the working-age population in October).
If, instead, we expand our view to all of those workers classified as “marginally attached to the labour force” — so, workers that are available and want a job, but have not looked for one in the past month; discouraged workers are a subset of this, as are those that have family responsibilities, ill health or a disability, or in school — that share of the population increases, but is still very small (1.9 million people or just 0.76% of the working-age population).
So, while changes in worker discouragement may have an impact on the participation rate on the margin, the change in the labour force participation rate is a reflection of structural shifts in the labour market that are driven by demographics more than anything else — this is the reason that the participation rate has remained (and will likely continue to remain) notably below its pre-recession peaks, plain and simple.
The first members of the “baby boom” generation — those born between 1946 and 1964 — starting turning 55 in 2001 and the ranks of the population aged 55+ has been rapidly increasing since, with a sharp acceleration seen start 2007 — the Census Bureau projects that we will see the population over 55 increase by two million per year out to 2020 (when the last of the baby boomers hit 55).
Now, if we look at the breakdown of the change in the population that is not in the labour force since the participation rate touched its recent in January 2007 (see chart below), it is perhaps unsurprising that the vast majority of these people falling out of the workforce are aged 55 or older (68% or 10.5 million people out of the 15.5 million increase in those not in the labour force).
Moreover, fully 93% of these people aged 55 or over that are now out of the labour force reported that they do not want a job now.
If this alone in not compelling enough that the odds are that these people are retirees and thus not likely to start looking for work again, let’s turn our attention to the data produced by the Bureau of Labor Statistics that looks at the flows of people within the labour market.
These data give an idea of the movements among the ranks of the “employed”, “unemployed” and “not in labour force” that makes up the headline employment numbers and what is of particular interest to us here is the number of people that remain outside of the labour force on a month-to-month basis.
The chart below showed that 93% of those classified as “not in the labour force” in October also resided there in September. Furthermore, since this share hit its cycle-low in June 2010, there has been a clear upward trend and we are currently around historically high levels.
In other words, an increasing share of Americans outside of the labour force are not making an effort to come back.
Now, as a fun exercise, let’s add all of these people over the age of 55 that do not want a job back to the labour force (there were 9.8 million more of them in October than back in January 2007) and look what that means for the participation rate.
The chart below plots the actual participation rate and what the rate would be if instead of leaving the labour force, those aged 55 and over that do not actually want a job remained in the workforce (either as unemployed or employed; for this context, it doesn’t matter).
What do you know? If we included these likely retirees in the labour force, the participation rate would have been 66.3% in October rather than the actual 62.4% — note that the pre-recession peak level was 66.4% in January 2007.
So, if not for these people who are very likely retirees leaving the labour force, the participation rate would effectively have remained steady at its pre-recession peaks — that increase in people aged 55 and over that do not want to work since 2007 has accounted for the entire four percentage point decline in the participation rate.
It is very simply the demographic impact of the baby boom generation reaching their golden years that is the story behind this drop in the participation rate and nothing else.
So, once again, any mention of the participation rate — or alternatively, mentions of a participation rate-adjusted unemployment rate or anything else of along those lines — as an indicator of labour market slack or as an argument for why the unemployment rate is obsolete and should be ignored, especially in the context of the policy discussion.