The Rise in Payroll Employment: James Hamilton Reads NRO Fuzzy Math
Lawrence Kudlow claimed:
Including revisions, January employment is a huge 317,000 above the initial December level.
James Hamilton reads this sentence and writes:
I’m not quite sure where Larry is getting this 317,000 figure, but I am inclined not to make too much of the data revisions. As the table on the right reveals, the revised figures released yesterday show 140,000 new jobs in December, rather than the 108,000 December job growth figure that was reported on January 6.
But let’s go to page 6 of the Employment Situation Summary that discusses revisions to establishment survey data:
In accordance with annual practice, the establishment survey data have been revised to reflect comprehensive universe counts of payroll jobs, or benchmarks. These counts are derived principally from unemployment insurance tax records for March 2005. As a result of the benchmark process, all not seasonally adjusted data series were subject to revision from April 2004 forward, the time period since the last benchmark was established. In addition, with this release, the seasonally adjusted establishment survey data from January 2001 forward were subject to revision due to the introduction of updated seasonal adjustment factors.
Had these revisions increased reported employment by something on the order of 93,000, then perhaps we might understand how Mr. Kudlow came up with his 317,000 figure. Now any reasonable person would say such playing games with numbers is rubbish, but note Kudlow loves to report the change in the reported household survey figures – never acknowledging the footnotes, which state “data affected by changes in population controls in January 2000, January 2003, January 2004, January 2005, and January 2006”. To illustrate how dishonest Kudlow et al. are being, let’s consider the following chart of the reported household survey figure from October 2002 to September 2003 – a period in which the employment to population ratio declined from 62.7% to 62%. Yet, the reported employment number appears to be higher on September 2003 than it was on October 2002 because of the jump in reported employment from December 2002 to January 2003. This jump occurred not because the employment to population ratio jumped. Rather it is related to the “changes in population controls in January 2003”.
Kudlow got the sign wrong per the recent revision in the payroll survey. Table C shows the December figure “as previously published” to be 134,468,000 as compared to the revised figure of 134,371,000. Now if some pessimist wished to be as dishonest as Kudlow has been in the past, he could not that the January 2006 figure of 134,564,000 is only 96,000 greater than the 134.468,000 December figure “as previously published”. But let’s not go down this dishonest path – January did see an increase in employment that was indeed almost 200,000, which was neither as explosive as Kudlow had claimed nor as small as his inappropriate reporting technique (done right) would have implied. The following graph shows reported employment from July 2005 to December 2005 “as previously published” as well as the current reporting from July 2005 to January 2006.
No wonder Dr. Hamilton was confused by Kudlow’s Kalculation – as Kudlow once again fails to tell his readers what he has done and this time fails to double check his math.