Jack Dunn - Reclaiming Common Sense

The Most Unloved Economic Indicator


The Weekly Unemployment claims data was released at 8:30 AM today, Thursday February 8th. The data released includes the non-seasonally adjusted (NSA) first-time unemployment (FTU) claims data and the Non-Seasonally Adjusted Continuing Claims (CC) data. The data that is reported, when it is reported, is the Seasonally Adjusted (SA) FTU and SA CC value. The seasonal factors used to convert the NSA data to the SA data changes from week to week, month to month, season to season, and year to year. When seasonally adjusted data from different weeks of the year are compared False Assertions Considered to be True (FACTs) are created. The biggest FACT was President Obama's Private Sector Job Creation streak. The second biggest FACT is the Seasonally adjusted First-time Unemployment Claims Streak. Both of these are covered in the article "Economic Urban Legends." What you need to know is that the trend for both NSA FTU and NSA CC data is downward through May. So what happened this past week?


The Non-seasonally Adjusted First-time Unemployment Claims data is at the Lowest it has ever been for the First-week of February. Ever. The unemployment claims program began during January of 1967.  Last year we had 259,000 NSA FTU claims. This year we had 240,636 claims. We often have had over 400,000 NSA FTU claims during the first week of February. How low can this number fall? Could it drop below 200,000 NSA FTU?


The Seasonally Adjusted First-time Claims data was "reported" at 221,000 claims. The world reported is in quotations because this number has been rarely reported since the election of 2016. If we used the seasonal factors used between 2000-2005 the headline number would have been reported as low as 205,000 claims. If the factor for the first week of February 1988 was applied to today's data then the number would have been reported at 201,000 claims. If we used the seasonal factors from the early years of the program then we would have had a headline number of 189,000 claims. This is fascinating how much this data is being actively ignored elsewhere.


The Non-Seasonally Adjusted Continuing Claims data for the fourth week of January was lower than the same week 1972 to date. This is not quite the lowest NSA CC level ever. The FTU claims data feeds into the continuing claims data. The CC data lags the FTU data by one week. A continuing claims level under under 3 million this time of year was good. We often have a continuing claims level under 4 million claims. There was the possibility of a minor uptick in the NSA CC value this week, based on prior year data. This was a drop of 2.9%. The continuing claims data should continue to fall through May.


How low can the Continuing Claims fall? There are weekly patterns that are observable in the data. Some weeks the NSA FTU claims pop up a little and the continuing claims follow higher the next week. Some weeks the NSA FTU drops and the NSA CC value falls the next week. The weeks where we traditionally see spikes are often seasonally adjusted into a smoother line. The authors of the report used to publish the seasonal factors months in advance. They no longer publish the data back to the beginning. They no longer archive the reports. The data indicates that we could see the lowest first-time claims data any given week of the year through October. Will we drop below the November 5, 1988 low this October or November? It is possible.


When will the NSA First-time Unemployment Claims level drop below 200,000? How low can the First-time Claims data fall?  We dropped below 200,000 NSA FTU for the first time "forever" during the second week of August 2015. We dropped below 200,000 NSA FTU three times during 2016, the weeks of 8/13, 9/10, and 9/24. we dropped below 200,000 NSA FTU claims four time s during 2017 between 7/29 and 8/26. Four times in a five week spread. We almost dropped below 200,000 on 9/30/2017 when we were at 204,000 NSA FTU claims. This information was ignored as it was happening. Could the election have turned out differently if the real numbers were reported? Will these numbers be referenced prior to the 2018 mid-term elections?


Next week we will receive the FTU claims data for the second week of February. This is important because the data from next week will give a glimpse to what is possible for the official U-3 unemployment number for the February Jobs report. The following week we will receive the continuing claims data for the same week for the month. The Current Population Survey data was strong this January. If you thought that the unemployment claims data for this month was great, the lowest since 2002. Will anybody else catch on to this trend. Spread the word.


It's the economy.