I welcome “abrupt policy adjustments”

Here’s an interesting report on unemployment:

The concurrent decline in labor force participation, however, has prompted many assertions that unemployment is falling “for the wrong reasons” — i.e., the unemployment rate is falling because unemployed Americans who can’t find work are becoming discouraged and dropping out of the labor force.

This idea has had profound implications for Fed policy.

Fed officials have sought to de-emphasize the decline in the headline unemployment rate — previously considered a key input to policy decision — suggesting it does not reflect the true health of the labor market. This orientation toward the labor market is being used as a justification inside the Fed for continued extraordinary monetary stimulus. 

Yet contrary to this popular narrative, the data suggest that the vast majority of the decline in labor force participation in recent years can be accounted for by the retirement of the “baby boomer” generation of American workers.

Consider the findings of a recent study by  Shigeru Fujita, a senior economist at the  Federal Reserve Bank of Philadelphia, titled “On the Causes of Declines in the Labor Force Participation Rate.”


Reasons for leaving the labor force

Federal Reserve Bank of Philadelphia, BLS

Reasons for leaving the labor force can be broken down into three categories: retirement, disability, and “other.” The “other” category is made up partly of discouraged workers. The chart illustrates that in recent years, retirement has been the primary driver of the decline in labor force participation.

Fujita demonstrates that “discouraged workers” only made up about a quarter of those leaving the labor force between 2007 and 2011, while “t he decline in the participation rate since the first quarter of 2012 is entirely  accounted for by increases in nonparticipation due to retirement.”

So far I am 100% on board with this “hawkish” article.  But then they begin to lose me:

If this is in fact the case, the current headline 6.7% unemployment rate may indeed reflect the true health of the labor market, the inflation the Fed is trying so hard to stoke may be a lot closer at hand than it expects, and the central bank risks falling behind the curve with regard to policy.

I see little risk of that happening, at least under the Evans Rule.  Remember, under the Evans Rule the Fed also looks at inflation.

The counterargument is that a wave of discouraged workers will re-enter the labor force as the job market improves, sending the labor force participation rate — and therefore, the unemployment rate — higher again.

Such developments would perhaps constitute a proper justification for continued zero-interest rate monetary policy, in line with the Fed’s projections, which imply no rate hikes for nearly two more years.

Nope, the real argument is that even under the assumption of this hawkish article we still have several millions of unemployed workers who need to find jobs.  There is still an output gap.  What’s their counterargument?

“Should we really be concerned about the Fed  being ‘behind the curve’ given that a rebound in participation should be  forthcoming with a rebound in the economy?” asks Drew Matus, deputy chief U.S. economist at UBS, in a note to clients.

Matus implies that the answer is yes.

“Although we believe that  there could be a modest cyclical rise in participation as the economy  improves, we believe the likely scale of the increase will not  significantly alter the basic equation,” he says.

“Payroll growth averaging 200,000  per month should continue to pull down the unemployment rate under all  but the most aggressive labor force expansion estimates.”

All of this suggests that labor market conditions are tighter than the popular narrative about the “accuracy” of the headline unemployment rate would have you believe.

If that is the case, the inflation that has thus far been elusive in this economic recovery may be closer than the consensus expects, especially if the unemployment rate continues to plummet toward the Fed’s estimated range of the unemployment rate in the long run between 5.2% and 6.0%.

The Fed should heed the message being transmitted by the headline unemployment rate if it wants to avoid the need for abrupt policy adjustments as the economy continues to improve.

We should welcome an “abrupt policy adjustment.”  It’s much better to get to full employment in one year and then abruptly adjust policy to keep NGDP rising along a 4.5% trend line, than it is to have a gradual recovery that asymptotically approaches full employment over many years.  Which has been the actual policy since 2009, in case anyone didn’t notice.

Despite my objections this is a very informative article, and should be read by all my fellow “doves” who believe there is a hidden army of 20 million unemployed just itching to get back to work.  There isn’t.  But 3 to 5 million excess unemployed for demand-side reasons is still a tragedy.  After we re-employ them we need to start working on the supply-side problems that reduce employment in America. Step one—replace the minimum wage with a wage subsidy.

PS.  Notice the retirement surge after 2011?  Count back 65 years and you get 1946.  Hmmm, what began in 1946?  For the foreseeable future we’ll have roughly as many workers retiring as entering the labor force.  (A bit fewer retiring right now, but a bit more than new entrants retiring in 10 years, ignoring immigration.)



22 Responses to “I welcome “abrupt policy adjustments””

  1. Gravatar of Wawawa Wawawa
    20. January 2014 at 08:41

    I’m still confused by the overall lack of attention given to LFPR for the prime working age group. I doubt its continued fall is driven by retirees.

  2. Gravatar of David Pinto David Pinto
    20. January 2014 at 08:45

    Does anyone else think the big rise in disabled was people finding a way to get money once they lost their jobs? Strange how it shadows the Other line.

  3. Gravatar of Randomize Randomize
    20. January 2014 at 08:51

    Yes, subsidize wages! The basic rules of supply and demand dictate that fixing labor prices above the market will cause a shortage of demand for labor. If people want more hours to go with their higher wages, roll out a subsidy instead of fixing prices!

  4. Gravatar of TravisV TravisV
    20. January 2014 at 09:13

    Dear Commenters,

    See Warren Buffett’s analysis of interest rates here:


    Does it actually make sense?

  5. Gravatar of Kevin Erdmann Kevin Erdmann
    20. January 2014 at 09:41

    Hear! Hear!

  6. Gravatar of ssumner ssumner
    20. January 2014 at 10:30

    Wawawa, Good point.

    David, Both liberals and conservatives agree that some of the rise in disability is related to unemployment. Not sure how much.

  7. Gravatar of W. Peden W. Peden
    20. January 2014 at 11:05

    “Remember, under the Evans Rule the Fed also looks at inflation.”

    That would be an improvement on current arrangements, whereby central banks look at inflation- if it’s too high. The Eurozone, but also it seems the US, has gone from inflation targeting (imperfect, but not that bad an idea) to having an inflation ceiling.

  8. Gravatar of Kevin Erdmann Kevin Erdmann
    20. January 2014 at 11:33


    If you’re defining prime working age as 25-54, then it is still demographics. LFP for 25-44 has been pretty level since mid-2011. The drop is all from 45-54 year olds, which I think is generally due to the bulk of baby boomers aging through that age group.

    The LFP of 45-54 year olds is lower than 25-44 year olds. And the LFP of 54 year olds is much lower than the LFP of 45 year olds. The baby boomer population bulge is deflating LFP for the 25-54 age group through both of these factors.

  9. Gravatar of TravisV TravisV
    20. January 2014 at 11:59

    Jon Hilsenrath:

    “Could Bernanke Land at Brooking After the Fed?”


    “The stars could be aligning for Federal Reserve Chairman Ben Bernanke to join the Brookings Institution after his term ends at the end of the month.

    Mr. Bernanke was the featured speaker at an event sponsored by Brookings last week, his last pubic appearance as Fed chairman. His friend and former deputy, Donald Kohn, has been a senior fellow at Brookings since leaving the Fed in 2010. And Brookings has recently established a new center on monetary and fiscal policy, called the Hutchins Center, with $10 million in funding from Glenn Hutchins, a founder of Silver Lake Partners, an investment fund. Mr. Hutchins also is a director on the board of the Federal Reserve Bank of New York.

    Mr. Bernanke plans to write books and give speeches after he leaves the Fed, and has said he plans to remain in Washington.

    Brookings, a Washington think tank, has been keen to land Mr. Bernanke and has had talks with him about it, said people familiar with the discussions. If he does plan to make a move there, he likely wouldn’t announce it until after his business at the Fed is finished. The Fed has a policy meeting Jan. 28-29 and his term ends Jan. 31.

    “There are lots of interesting things to work on and I look forward to writing and speaking and taking a little bit more time to contemplate some interesting issues,” he said in November at the National Economists Club annual dinner.

    A spokeswoman for the Fed declined to comment. A spokeswoman from Brookings didn’t respond to requests for comment.”

  10. Gravatar of benjamin cole benjamin cole
    20. January 2014 at 12:35

    Excellent blogging.
    Couple cavils.
    Immigrants can and do add to labor force, dampening already feeble inflationary trens.
    People can unretire or consult or take part-time jobs. I know a retired factory inspector at Merck makes $60k inspecting a few factories a year oncall.
    There is a deep-seated phobia and fixation among some regarding inflation. They just have been driven nuts by QE but dead inflation

    The truth is the Fed has done too little.

  11. Gravatar of steve from virginia steve from virginia
    20. January 2014 at 13:39

    Hard to have increased jobs when retail is taking it in the neck:


    Reduced shoppers => less credit => declining money velocity => shrinking M3 and the trend toward deflation.

  12. Gravatar of Wawawa Wawawa
    20. January 2014 at 14:21

    Kevin Erdmann:
    I’ll have to dig into the data a bit more, I’m assuming your comments are empirical in nature. Anecdotally, I find it difficult to believe that we are experiencing a large number of 45-54 yo going into early retirement after (presumably) recently experiencing such huge losses in their retirement accounts. Or maybe recent market rallies have made this group content with getting back to par and happy to cash out & retire to avoid further losses?

  13. Gravatar of Edwin Alvarenga Edwin Alvarenga
    20. January 2014 at 14:35

    I’m confused. Then why is the participation rate actually rising for for 55 and older?
    The same holds for 65 and older.

    25 to 54 seems is the only demographic that the participation rate falling for.

  14. Gravatar of Don Don
    20. January 2014 at 14:45

    It is hard to parse the graphs without specific criteria for states of “retired”, “disabled”, and “other”. I can guess “disabled” means collecting SS Disability, but when exactly does “other” become “retired”?

  15. Gravatar of benjamin cole benjamin cole
    20. January 2014 at 16:52

    There are nearly 12 million Americans now collecting “disability” monthly checks, 3.7 million of them vets, more than 8 million SSDI…
    Probably disability rolls should be cut in half…for starters…

  16. Gravatar of Kevin Erdmann Kevin Erdmann
    20. January 2014 at 18:52

    Wawawa & Edwin,

    That’s what makes these numbers so tough – there are so many moving parts. With something this complex, the public consensus tends to be swayed more by predestined narratives. So, it is common for all the odd movements to be blamed on the business cycle. The business cycle can move LFP, but not that much.

    Here are some of the moving parts that I see that are affecting all the groups you mention:

    Baby boomers are working later in life for cultural reasons, and because they are healthier at an older age than their parents were.

    Young women started entering the labor force at a higher rate in the 1970s & 1980s. For most ages, female LFP peaked 10 or 15 years ago. But, for the oldest age groups, there is probably still some catch up happening on the female side, so for the 55+ set, I think you’ll find that most of the growth is from the female side.

    Even though baby boomers are working later, they are still getting older, so more 60 year olds are working today than 60 year olds were working 10 years ago, but there are a lot less 60 year olds working today compared to when they were 50 year olds 10 years ago. So, even though boomer LFP is growing within age groups, the overwhelming factor is that they are moving through the age groups where a lot of people start leaving the labor force. I hope that makes sense.

    There has been a 50 year or more slow downward trend in LFP among the prime working ages. That was hidden by the fact that women upped their LFP in the 70′s and 80′s. It looks like we had a big hump in the 90s and now something is broken. If women hadn’t had the one-time pop in LFP, it would be a very long term slightly downward sloping, fairly straight line with a little 2% high hump in the 1990′s when boomers were at the prime working age.

    Here is what long term LFP would look like if women didn’t have the one-time increase in LFP:

    Here’s a post where I summarize some of the LFP movements by age & gender:

  17. Gravatar of ssumner ssumner
    21. January 2014 at 05:47

    Kevin, Thanks, That’s very informative.

  18. Gravatar of pct pct
    21. January 2014 at 14:06

    I doubt both that the retired/other disjunction is as clean cut as one might think, and that discouraged workers all fall into the “other” category. Many boomers planned on working past 65, particularly ones whose portfolios were devastated in 2008. However, someone who is over 62, just laid off, and aware of the odds against his getting a new job at that age could easily give up, start collecting SS, and count as retired, no matter how discouraged he might be.

  19. Gravatar of flow5 flow5
    21. January 2014 at 14:41

    “We should welcome an “abrupt policy adjustment”

    This is absolutely the most significant comment you’ve ever made Sumner. That’s the essence of monetary policy, its immediate impact. I don’t know how to estimate “output gaps”. But I can tell you in advance when the Fed is going to screw up. And as soon as the FOMC realizes they’ve errored, I’ve watched them act quickly & try & cover them up (See Bernanke’s Feb 27th testimony to Congress). So you can’t re-create a time series to prove this.

    Every miscalculation since the Great-Depression (all the boom/busts), were predictable & were preventable.

    Greenspan’s still trying to figure out “Black Monday”, let alone the Great-Recession. (See “The Map & the Territory”)

    All the dislocations were all the same. I predicted the unpredicable (the 2010 May 6th “flash crash” 6 months in advance & within one day). I sent an e-mail to one of the Fed’s senior on March 30, 2010 detailing the trajectory. Stocks fell as liquidity dropped. The “trading desk” intervened to the extent that the drop was rendered undetectable ex-post. My point is that monetary policy is correctable in a very short time frame. And that the quicker the FOMC acts/corrects, the less the damage.

  20. Gravatar of Tom Brown Tom Brown
    21. January 2014 at 16:23

    Scott, flow5 above mentions “all the boom/busts” but I recall you writing once that booms and busts will still exist even with good monetary policy. You made a distinction between a “bubble” and a boom. Am I recalling that correctly? Can you elaborate?

  21. Gravatar of flow5 flow5
    23. January 2014 at 08:40

    No, the business cycle can be completely eliminated. And you can only achieve “abrupt policy adjustments” by targeting legal reserves (i.e., money flows). The Fed CANNOT execute abrupt changes using interest rates as the monetary transmission mechanism. Keynes’ liquidity preference curve is a false doctrine.

  22. Gravatar of ssumner ssumner
    24. January 2014 at 06:53

    Tom, Asset prices would still have big cycles. Output would have some variability, but less than now.

Leave a Reply