US LABOR MARKET: Massachusetts Lessons For Texas Jobless Claims Fraud

With the rise in Texas initial jobless claims directly linked to ID fraud, we revisit the Massachusetts example from back in May 2023 which saw subsequent revisions to the series two weeks after being first officially recognized. These revisions came outside of the regular claims publication (albeit only by a few minutes) which makes it hard to predict when we might hear something for Texas. Tomorrow’s claims report will no doubt see particular focus on this front but the issue could remain outstanding. 

• As noted yesterday, the Texas Workforce Commission indicated that the rise in initial claims was “directly related to an increased volume of fraudulent claim attempts… Since Labor Day, we’ve observed an uptick in ID fraud claim attempts” (cited by Axios, link).  
• Recall that last week’s jobless claims release surprisingly jumped to a seasonally adjusted 263k (cons 235k) as the non-seasonally level of claims in Texas increased 15.3k vs 7.9k nationally.  
• Axios added that: "The Labor Department did not respond to questions about whether Texas communicated that the fraud issue inflated its claims - and if so, why it was not flagged in the public release." We aren’t surprised by the latter, with it taking some time for fraud in Massachusetts in 2023 to first be acknowledged and then corrected.
• Back in May 2023, national initial jobless claims surprisingly jumped to a seasonally adjusted 264k (cons 245k) in the week to May 6, with Massachusetts accounting for 6.4k of the 14k increase in the national non-seasonally adjusted level as MA continued some sizeable weekly increases. It saw the outright level increase to 34.9k (second only to California’s 46k despite a population less than a fifth of its size).
• The next day (May 12 after the May 11 publication), MA said that the increase was due to fraud attempts.
• However, it wasn’t until May 25 that MA released significant revisions just a few minutes before the publication of the regular claims release. What was seen as the peak 35k initially (before a 14k drop to 21k the week after) was subsequently revised down to 5.5k (and 3.8k).
• Sizeable revisions went back almost three months, although when it comes to Texas, the explicit reference to Labor Day suggests a more limited lookback to the roughly three weeks to Sep 1. Another difference this time is the sudden pop higher for Texas vs the more concerted build in MA.