Tuesday, December 17, 2019

Employment Cheer at Christmas?

I wanted to write something about December and Christmas and as usual I wanted to use macroeconomic data to illustrate something -- this time something about the US economy during the holidays. The obvious thing is that we spend a lot of money for gifts at the end of the year. Ho Ho Ho.

In looking at monthly spending series, I wasn't happy with what was available so I decided to try monthly employment. I wanted to better understand what happens to US employment over the holidays. The first thing I did was chart the numbers, and that chart is found below. It is hard to see end of year employment patterns on a graph like that, but it does underscore the dramatic and continuous growth of employment over a long period of time.

It is easy to see the many times when one month's employment fell, but the thing that screams at you from the graph is how persistently employment in the US has increased since the days before World War II. We've gone from about 30 million jobs to almost 160 million. That's quite a streak.

If you look at just the last 21 years -- employment went from 129 million at the end of 1998 to about 154 million at the end of 2019. That's a rise of about 25 million jobs in those years.

But that is not what I wanted to write about today. Hidden in all those data points is what usually happens in December of each year. The table below the chart shows you some information about monthly changes.

I will summarize it here.

Among the information collected from the Bureau of Economic Analysis:

          Employment rose each month, on average by about 390,000 jobs from 1998 to 2018. Multiply that number by 12, and you get how much employment increased over the course of a whole average year from 1998 to 2018.

          With respect to the end of each year, the average change in each month over those 21 years was:
          November      294,000 per month
          December      -214,000
          January       -2,860,000
          February         729,000
          March             705,000

If you take those five months as a block, US employment averaged a decline of about 1.4 million jobs over those five months each year. But all the job loss was found in December and January of each year.

The wonderful holiday season with all that spending found employment drastically declining in December and January of each year. Even if you add in the strong snap backs in the following February and March (and the increase in November), you still get a strong decline for those five months on average each year.

Isn't that crazy! Each year jobs rise by an average of 390,000 each month despite the fact that in those five winter months employment declines by 1.4 million jobs. Those other seven months carry quite the employment load.

Looking at individual months supports the general view. In 2017-18, for example, employment change in the five winter months was -832,000 while the change from November to November was 1,265,000. Similarly in 2018-19, the numbers were -989,000 and +1,481,000.

When thinking of why this occurs, many things come to mind. We call these months holidays for a good reason. People often take time off from employment during these months. It might also be true that it takes employment in production in the months before December to create all the goods that will get sold in December and January. So production and employment cycles may be geared to working harder in the months before the holiday.

Finally, I purposely used data that was not seasonally adjusted. Much of what we read about has already been smoothed by seasonal adjustment factors. I wanted to see the real swings in the data.

What else could explain these crazy swings in employment in the USA? I am no expert on monthly employment changes, and even with the above reasoning, it blows me away to see these wild changes.





FRED Graph Observations
Federal Reserve Economic Data
Link: https://fred.stlouisfed.org
Help: https://fred.stlouisfed.org/help-faq
Economic Research Division
Federal Reserve Bank of St. Louis
PAYNSA All Employees, Total Nonfarm, Thousands of Persons, Monthly, Not Seasonally Adjusted
Frequency: Monthly
observation_date PAYNSA
Employment  CHG
1998-11-01 128290 403
1998-12-01 128459 169
1999-01-01 125708 -2751
1999-02-01 126696 988
1999-03-01 127409 713 -478 2027
1999-11-01 131510 461
1999-12-01 131646 136
2000-01-01 129005 -2641
2000-02-01 129667 662
2000-03-01 130764 1097 -285 2209
2000-11-01 133614 365
2000-12-01 133555 -59
2001-01-01 130681 -2874
2001-02-01 131348 667
2001-03-01 131942 594
2001-11-01 132155 -189
2001-12-01 131773 -382
2002-01-01 128890 -2883
2002-02-01 129362 472
2002-03-01 129969 607
2002-11-01 131667 121
2002-12-01 131259 -408
2003-01-01 128577 -2682
2003-02-01 128994 417
2003-03-01 129490 496 -2056 -1068
2003-11-01 131565 164
2003-12-01 131385 -180
2004-01-01 128726 -2659
2004-02-01 129337 611 -2064 -153
2004-03-01 130378
2004-11-01 133649 251
2004-12-01 133418 -231
2005-01-01 130713 -2705
2005-02-01 131536 823
2005-03-01 132376 840 -1022 882
2005-11-01 136174 557
2005-12-01 135973 -201
2006-01-01 133320 -2653
2006-02-01 134245 925
2006-03-01 135226 981 -391 1637
2006-11-01 138234 385
2006-12-01 138124 -110
2007-01-01 135334 -2790
2007-02-01 136026 692
2007-03-01 136923 897 -926 783
2007-11-01 139510 312
2007-12-01 139297 -213
2008-01-01 136268 -3029
2008-02-01 136787 519
2008-03-01 137378 591 -1820 -352
2008-11-01 136761 -681
2008-12-01 135732 -1029
2009-01-01 132042 -3690
2009-02-01 131808 -234
2009-03-01 131675 -133 -5767 -6239
2009-11-01 131236 48
2009-12-01 130690 -546
2010-01-01 127820 -2870
2010-02-01 128255 435
2010-03-01 129089 834 -2099 -2770
2010-11-01 131947 306
2010-12-01 131641 -306
2011-01-01 128778 -2863
2011-02-01 129592 814
2011-03-01 130499 907 -1142 287
2011-11-01 133893 335
2011-12-01 133718 -175
2012-01-01 131113 -2605
2012-02-01 132067 954
2012-03-01 132971 904 -587 1256
2012-11-01 136039 390
2012-12-01 135964 -75
2013-01-01 133081 -2883
2013-02-01 134120 1039
2013-03-01 134918 798 -731 1064
2013-11-01 138543 524
2013-12-01 138292 -251
2014-01-01 135488 -2804
2014-02-01 136229 741
2014-03-01 137187 958 -832 1264
2014-11-01 141331 465
2014-12-01 141326 -5
2015-01-01 138511 -2815
2015-02-01 139343 832
2015-03-01 140099 756 -767 1774
2015-11-01 144066 421
2015-12-01 144063 -3
2016-01-01 141088 -2975
2016-02-01 141919 831
2016-03-01 142814 895 -831 1531
2016-11-01 146482 433
2016-12-01 146270 -212
2017-01-01 143393 -2877
2017-02-01 144423 1030
2017-03-01 145078 655 -971 1169
2017-11-01 148774 574
2017-12-01 148526 -248
2018-01-01 145428 -3098
2018-02-01 146665 1237
2018-03-01 147368 703 -832 1265
2018-11-01 151375 522
2018-12-01 151203 -172
2019-01-01 148295 -2908
2019-02-01 149148 853
2019-03-01 149864 716 -989 1481
2019-11-01 153624 622
November 294
December -214
January -2860
February 729
March 705 -1073 390
Total Nov to Mar -1413.6

No comments:

Post a Comment