Jack Dunn - Reclaiming Common Sense

Another Thursday - Another Unemployment Claims Report


The weekly unemployment claims report used to be breaking news at 8:30 AM EST every Thursday that it was released.  The headline number is the seasonally adjusted (SA) first-time unemployment (FTU) claims data. If it was under 300,000 "all things were right with the world." It was so important that the Obama Administration manufactured a FACT (False Assertion Considered to be True) regarding consecutive weeks with a number under 300,000 claims. This is one of the economic urban legends created under the Obama Administration. A second number that would receive a nod, from time to time was the seasonally adjusted (SA) continuing Claims (CC) data. A number under two million was considered good. The problem is that the seasonal factors that are used to convert the non-seasonally adjusted (NSA) data to the SA data change by category, week, month, season and year. Comparing SA FTU data from different seasons created some seriously misleading FACTs. The NSA FTU number may fall below 175,000 this August or September. It is possible that the NSA CC value may drop below 1.5 million claims by October. The overall trends are heading downward.  What happened this week?


The Non-seasonally adjusted first-time claims data dropped to under 200,000 claims, again. We have rarely had fewer than 200,000 NSA FTU claims since 1974. We broke below 200,000 during the fall of 2015. We did it again during August of 2016 and August of 2017. Normally the low of the year is during the final week of August of the final week of September. We are approaching the August low of the past few years during April. The trend is still downward. Watch the bottom line of the trend pitchfork. Last week it was proposed that we could drop below 175,000 NSA FTU claims by September. Could we hit it by the end of June? That would be comparable to 1973, and we have been consistently beating the 1973 weekly values this year.


The seasonally adjusted first-time claims data  could have been reported under 175,000. Look at the data for April 21, 2012. We had nearly double the NSA FTU claims. If we used the same seasonal factor as that year then we would have seen a lower value reported than was reported. If we used the seasonal factors from 1967-1972 the SA FTU value would have been reported under 180,000 and as low as 168,000 claims. If we are going to make comparisons to seasonally adjusted values then we need comparable data, from comparable seasons and weeks, with comparable seasonal factors.


The non-seasonally adjusted continuing claims data dropped again! This makes sense. First-time claims are falling. First-time claims feed into continuing claims. We have 174,000 fewer NSA CC this week than last year at this time. It now seems very likely that we will drop below 1.5 million claims by the first week of October. Could we drop below 1.25 million claims by the first week of October. If the continuing claims data follows a similar pattern as the FTU claims data, the 1.252 million claim level seen during 1973 is within reach.


The seasonally adjusted continuing claims was over-reported at 1.836 million claims. Using "any" seasonal factor used during the second week of April, the SA CC value would have been reported lower than it was reported. Could it have been reported under 1.7 million? You bet.


The weekly unemployment claims data make a difference. The Weekly claims data is a separate database from the Employment Situation CPS data for unemployed workers. Right now the difference between the NSA CC value for mid-March and Mid-April  is down 243,000 claims. Could this mean that we have a huge jobs number? It is very possible. The crunching of the data for next week's April Jobs report will begin in earnest tomorrow, after the GDP report is released. This was a solid, almost historic, unemployment claims report. It is unloved.


It's the economy.