Ed Dolan's Econ Blog

Forget the Weak Payroll Numbers. January’s Employment Report was Fundamentally Positive

Although the news that the U.S. economy generated just 113,000 new payroll jobs in January 2014 disappointed many observers, the latest report from the BLS on the employment situation was fundamentally positive. That was evident not only from the 6.6 percent unemployment rate, down nearly half a percentage point over the last two months, but also from many underlying measures of employment stress—part-time work, long-term joblessness, and others.

Let’s start with the bad news and get it out of the way. January’s 113,000 new payroll jobs marked the second month in a row of low job growth. December’s even lower figure was revised up by just 1,000 jobs to 75,000. Even here, though, the news was not all bad. The relatively robust November job gain was revised up from a first-reported 203,000 to 274,000. In addition, the BLS rebenchmarked its data, as it does each year, to reflect a more comprehensive count of payrolls. The rebenchmarking increased job growth for the year by 136,000, bringing the total gain to 2,322,000. The following chart shows the rebenchmarked data.

Data from the household survey were considerably more upbeat than those from the establishment survey on which the data for payroll jobs are based. The two surveys differ in several ways. Among other things, the household survey includes self-employed and farm workers. It also counts workers, not jobs; one worker with two jobs gets double-counted in the establishment survey.

The household survey showed strong improvement in the labor market—630,000 more employed workers and 115,000 unemployed than in December. The unemployment rate fell to 6.6 percent, the lowest in more than five years.  In contrast to December, when a decrease in the unemployment rate was largely attributed to a decrease in the labor force, the number of people working or looking for work rose by 523,000. Both the labor force participation rate and the employment-population ratio increased.

In addition to the standard unemployment rate, the BLS reports a broader measure of labor market stress. That indicator, known as U-6, includes discouraged workers and involuntary part-time workers as well as the officially unemployed. U-6 decreased sharply in January to 12.7 percent of the labor force, its lowest level in more than five years. The next chart shows both standard and broad unemployment.

The drop in U-6 was caused, in part, by a sharp decrease in the number of people involuntarily working part-time, or, to use the official terminology, working part-time “for economic reasons.” Those include people whose hours have been cut because of slack business conditions and people who would prefer a full-time job but can only find part-time work. As the next chart shows, involuntary part-time workers decreased from 5 percent of the labor force to 4.7 percent, the lowest since late 2008. At the same time, there was an increase in the number of people voluntarily working part-time. In all, nearly three-quarters of people who worked part-time were doing so for “noneconomic reasons,” such as childcare, family obligations, school, or retirement. (For more details on the evolution of voluntary and involuntary part-time work in the United States, see this recent post.)

Finally, the percentage of the unemployed who had been out of work for 27 weeks or longer also dropped to a new low for the recovery. Long-term unemployment has been unusually high throughout the recession. The January figures represent a welcome resumption of a downtrend that had been interrupted in November and December. The median and mean duration of unemployment both decreased in the month.

On the whole, then, the employment situation showed continuing improvement in January. Despite a disappointing number of new headline jobs, several data series hit new lows for the recovery, including both the standard and broad unemployment rates, involuntary part-time work, and long-term unemployment. The civilian labor force, the labor force participation rate, and the employment-population ratio all increased. These developments point to continuing expansion of the economy at the beginning of 2014.

17 Responses to “Forget the Weak Payroll Numbers. January’s Employment Report was Fundamentally Positive”

windrivenFebruary 9th, 2014 at 1:59 pm

Dr. Dolan, I would agree that the continuing improvement in U6 is encouraging but I don't see any meaningful improvement in participation rate ( which has been in decline since 2007-2008. Some of this can be attributed to Boomers retiring but certainly not all.

I serve on the board of directors of an NGO focused on workforce issues in the Pacific Northwest and from my vantage point the labor market still looks very soft.

Ed Dolan EdDolanFebruary 9th, 2014 at 8:37 pm

I agree that the labor market is still soft. The question is how fast it is firming up. Part of the answer lies in how much of the decline in the employment-population ratio (and the related indicator, the labor force participation rate) is due to our aging population. The NY Fed recently put out a study that breaks out the demographic component. It indicates that more than half of the decline in the EPR since 2007 is due to demographics. You can read their basis for that finding here:

windrivenFebruary 11th, 2014 at 12:16 am

You'll forgive me for noting that the actual E/P recovered nicely after each of the prior recessions while it is flat as a dead man's EKG after this one. The 'predicted' and the actual look set to converge around 2015 proving, I guess … something. Maybe that if you wait long enough the model will intersect with reality.

I don't want to sound like a crank but much of the research coming out of the Fed looks for all the world like Officer Barbrady shooing people off saying, "move along, nothing to see here."

I'm old enough to remember each of the recessions depicted on the graph. This one feels different. Different in kind rather than degree. Wealth has shifted. The meaning of employment is changing. People that I know and people that I meet, many of them anyway, have lost confidence that the future holds broad promise for them or for their children. Very disturbing.

Ed Dolan EdDolanFebruary 11th, 2014 at 9:56 am

"This one feels different. Different in kind rather than degree"

Of course it is different, and different in many ways. However, it is indisputable that one of the differences is that the population is aging to a degree never seen in human history, let alone one lifetime.

The Fed is not the only one to note that changing demographics impacts the E-R ratio. Plenty of respectable private economists agree.

windrivenFebruary 11th, 2014 at 12:38 am

Here is a chart from FRED that puts the lie to the Kapon-Tracy paper.

The blue line (55 and older) shows growth in employment throughout the period in question while 25-54 (red line) looks suspiciously like the E/P actual. Hmmmmm…

– Thanks to commenter CEDO on the Kapon-Tracy page for bringing the chart to my attention.

Ed Dolan EdDolanFebruary 11th, 2014 at 10:17 am

Thanks for steering me to this chart and the comment thread at the Kapon-Tracy paper. I read the thread and looked at the charts. What I don't understand is why you think this comment (Posted to K-T by "Chris") does not resolve the issue."Can you please explain what I am missing here?

Here is the comment by "Chris":

"for all those that chart the participation rate of 24-54 year olds then 55+ year olds:

If i have a participation rate of 20% for over 55s and they are 10% of the population, what happens when that participation rises to 25% and they are 30% of the population?

The overall participation rate of 65% (say thats what it is) will have to fall despite the rise in the 55+ participation rate.

eg 60/100 people work … participation of 55+ is 10% and they are 25% of population (25) so only 2.5 65+ workers … then participation rises to 20% and ratio rises to 40% (40), then there would be 8 workers … either participation of the remainder would have to rocket higher or overall participation would fall."

windrivenFebruary 11th, 2014 at 11:52 am

Dr. Dolan, with all due respect Chris's comment is silly.

S/he says: "The overall participation rate of 65% (say thats what it is) will have to fall despite the rise in the 55 participation rate."

????? The overall participation rate remains unchanged. It is what it is. The demographic composition might change but the participation rate is the same. The denominator is fixed. More people participating from one demographic would mean fewer from another but the composition doesn't affect the total participation rate. The numerator also remains, more or less, the same. The balance of demographics in the numerator would change but not the total value.

What Chris's point should have been is that the slack participation rate appears to be differentially more burdensome to younger workers. That seems to be the message of the FRED chart.

Ed Dolan EdDolanFebruary 11th, 2014 at 8:24 pm

Maybe Chris's numerical example wasn't clear enough. Try these simpler numbers:

Year 1: 100 people in population, of which:
30 are 16-24 with 40% participation, so 12 in labor force
50 are 25-54 with 80% rate, 40 in LF
20 are 55+ with 20% rate, 4 in LF
Total 56 in labor force, LPR 56%

Year 2: 100 people in population, of which:
20 are 16-24 with 40% participation, so 8 in labor force
40 are 25-54 with 80% rate, 32 in LF
40 are 55+ with 25% rate, 10 in LF
Total 50 in labor force, LPR 50%

The point is, if the population ages, you can get a decrease in LPR even if the LPR of the oldest group goes up, and even if the LPR of the youngest group does not change. All K-T do is give a more precise calculation based on real numbers rather than the made-up numbers above, but the concept is the same.

windrivenFebruary 12th, 2014 at 12:39 pm

Thank you for your kind reply Dr. Dolan.

I understood Chris' numbers and I understand yours. We all agree on the arithmetic. It is the underlying argument that is flawed and perhaps I was inarticulate in my criticism.

Strip away all of the numbers and Chris' argument is banal: if you reduce the number of employed while holding the total population constant the labor force participation rate will fall. All those participation percentages by age cohort amount to nothing more than Three Card Monte. We are living in a job starved economy; there are more people who want jobs than can get them. In Example A there are 56 in the workforce and in Example B there are 50. In an economy with very low unemployment one could mount this argument as the population aged. But for this argument to make sense in the context of the FRED chart there would have had to have been a sharp reduction in the 25-54 population cohort between the years 2007 and 2011 or a very large increase in the non-working 55+ cohort. I have seen no evidence of that, have you?

The FRED chart clearly shows a sharp drop in the 0-55 demo of those "Employed, Usually Work Full Time" from ~.35 to ~.31. That amounts to an 11% decline. For Chris' argument to make sense the proportion of Americans in that demo would to have seen a similar decline. That is manifestly not the case.

Chris' argument is perfectly sensible in the abstract. It is the application of that argument to the current economic and population realities that is, IMHO, bogus.

Ed Dolan EdDolanFebruary 12th, 2014 at 6:43 pm

Thanks for the clarification. I'll look into it.

One comment: You should not forget that the number of jobs offered in the economy is not independent of the number of people who want to work. For example, if there were a culturally or demographically driven change in preferences toward more leisure or earlier retirement, the eventual result would be a lower level of potential real output, and fewer jobs when the economy was operating at that potential. I do not mean to suggest that the economy is not now below potential, but even if it never strayed from potential, the number of jobs would depend both on supply and demand.

windrivenFebruary 9th, 2014 at 2:06 pm

As an aside, it would be really enlightening if BLS or some other authority tracked total hours paid every month. This, it seems to me, would be a valuable measure of economic health as it ignored part v full time; in fact it ignores unemployment completely. It simply says workers were paid for this many hours last month versus this many hours this month.

metingun2February 11th, 2014 at 6:22 am

In fact what really important is `the direction of both curves in parallel and downwardly ` and the gap of roughly % 6 in between the two curves is the result of ` incurable social reasons behind the unemployment ` like someone does not like the `wide spectrum of ever available vacant jobs offered out there` and does not want to participate and contribute in the society in positive ways ; for these category U-6 unemployment data there is no miracle cure in any economy in this planet earth , so the gap in between those two parallel curves will always persist , unless they are given somehow `extra incentive to work` for the betterment of their daily lives on the other hand – may be most of them are happy with their lives what they are “not doing anything” statistically recordable — . Please cont. Part 2

windrivenFebruary 11th, 2014 at 4:36 pm

"like someone does not like the `wide spectrum of ever available vacant jobs offered out there and does not want to participate and contribute in the society in positive ways"

Did you intend this ironically?

I've read all three parts. I usually think myself a clever enough guy. Maybe not. I have no freaking idea what your point is. Is it along the lines of: those who ain't working are … whatever politically correct version of deadbeat you prefer? If so why did it take three comments to get it out?

Or is this all intended to be tongue in cheek? I don't know. Your comment is such a rambling melange of jumbled thoughts and disconnected ideas that it leaves me lost as a blind dog in a meatlocker.

metingun2February 11th, 2014 at 6:22 am

Part 2.
Federal Reserve policies are of course designed and expected to follow the `bottom curve` as always with no exception ; and as far as Fed Reserve goes unemployment rate is % 6.6 as is , Fed Reserve policy makers simply disregard the top curve ; and if you really want to know for the Economical reasons what the next `Fed ` move is and to predict the future money policy decision by the Federal Reserve just follow the bottom curve . Of course unless you want to approach the issue as a `social worker` or may be coming ` November 2014 election policy designer to” exploit the irrelevant information” for potential undecided prospective voters for the benefit of hopeless Tea Party guys who had shepherded and coned the Republicans to shut down the Federal Government and even came too close to default US Dollar the good faith and standing of World`s Reserve Currency . Please cont. to bottom line .

metingun2February 11th, 2014 at 6:23 am

The bottom line is ; ` those young `able bodied , presumed healthy` guys/girls who persistently belong to the top curve so called U-6 unemployment statistics year after year , and obviously they are so reluctant to pic any `job` available out there , which does not require a college degree or skill or even any education whatsoever ; they have to learn to be flexible in life and to help themselves ; because no body else can help them ; `if they do not want to participate and be flexible ` I guess at the end of the line that is nobodies business to question their motive since it is `free country` after all . By the way ; It is always good to say `the glass is half full ` rather than `half empty .

benleetFebruary 18th, 2014 at 2:05 pm

The LFPR is at 63.0, a low that last happened in April 1978, 36 years ago almost. Add the non-participants from the high mark of participation, Jan 2000, when it LFPR was 67.3%, and you add another 10.6 million to the official 10.2 million, the U3 rate reaches 12.5%. Heidi Sheirholz at EPI publishes a LFPR adjusted unemployment rate now, it's 10.2%. I think she's underestimated the actual drop-outs and it should be 11.7%. Anyway, and whole article about the improving state of the labor market should include the above fact about not since 1978 has LFPR been so low. I looked at the OECD rates recently, I think Norway and Denmark and Germany were the highest, Turkey, Mexico, Poland and poor South Africa were the lowest. Take that as an indicator. We are in a recession still — don't be misled or misleading.