Perspectives

Progress

Get Insights in Your Inbox

shutterstock 795017047 scaled

As we have been saying, the seasonal adjustment process for initial jobless claims has been wholly inappropriate given the scale of job losses during the pandemic. Because the process is multiplicative rather than additive, the adjustment factor has now inflated the actual claims numbers by over 4.6 million since March, grossly exaggerating the true magnitude of job loss. We’re happy to report that the Bureau of Labor Statistics has realized the error of this approach, and announced that they will switch to an additive seasonal adjustment method starting in September. Clearly this discrepancy had grown too large for anyone to ignore and we’re glad to see a more rational approach to measuring this one piece of economic data.

Stephen Douglass

Stephen Douglass

Chief Economist

Stephen analyzes the global economy, monetary and fiscal policy, as well as interest rate markets. He is a leading voice on economic, market and portfolio matters—engaging…

Share

Partner With NISA

Discuss Your Portfolio Objectives with Us

Insights to Your Inbox

Receive timely updates from NISA delivered to your inbox. Select the topics most relevant to your interests.

Topics of interest(Required)

We respect your inbox. Unsubscribe anytime.

Agree to Terms

Please review and accept the following to proceed.

I have read and agree to the Terms of Use, Disclaimer, PSRX Disclaimer and Privacy Policy. I am either (i) an investment professional and an employee of an institutional investor, or a consultant to an institutional investor, or (ii) an employee of, or a student in, an institution of higher learning and I am involved in the study, research or teaching of subjects related to investments, finance, or economics. I reside in the United States or Canada. I understand that the information is not and should not be regarded as investment advice or as a recommendation regarding a course of action.

By clicking “Accept” below, you hereby acknowledge, understand and accept the foregoing.