Initial Jobless Claims

Initial jobless claims is a primary leading macroeconomic indicator. The data here describe the number of first time or initial claims for unemployment benefits statewide. The underlying theory is that a stronger economy will lead to more people employed, thus the number of people that are newly unemployed should drop. As a macro variable, it represents emerging unemployment and signals what is likely to happen to the number of individuals that will continue to file for unemployment benefits (continued claims). The numbers are released on a weekly basis by the Department of Labor.

Initial Unemployment Claims

Initial unemployment benefit claims are plotted in blue1. We can see that exiting the 1980’s first time claims dropped off slightly, reverting closer to the mean, for all of the 1990s and the first half of the 2000s. Hurricane Katrina made landfall in August 2005 and this led to a relatively large, yet temporary, increase in initial jobless claims start in September 2005. By January of 2006 the number of first time jobless claims was even below average. Coinciding with the stock market collapse in September 2008, initial claims spiked to almost four times the long-run average. This spike was also transient, reverting back to its mean the next month.

The average and median claims are 3833 and 3379 respectively. The higher average indicates that the sharp spike in claims during 2005 and 2008, which are clearly outliers, is skewing the average upwards.

Forecasts of Initial Jobless Claims

Forecasts2 for Louisiana initial claims are made from an aggregation of multiple statistical models designed to approximate the underlying data generating process of the available data. A Bayesian model averaging approach is used here to capture the joint uncertainty that any given model may be misspecified as well as capturing the probabilistic uncertainty inherent in each individual estimate. Observed data is given in blue while forecasts are presented in orange. The weighted average of all models used is represented by the solid orange line. The upper bound and lower of the cone of uncertainty surrounding these estimates is represented by the dashed upper and lower lines respectively.

 


  1. Please note that the graphs below are interactive HTML widgets. Please hover over each to examine the underlying data that comprises them.

  2. Forecasts are provided as a convenience and for informational purposes only without any explicit or implied warranty. The authors and publishers of this post and site bear no responsibility for the information provided here and cannot be held liable for any negative consequences that may arise due its publication. Forecasting the future is inherently a tricky proposition, and all forecasts have an error term attached to them. Please exercise caution when making financial and business decisions based on the information provided. Use this information as a single input into your decisions making process.

Patrick is an assistant professor of Economics at Louisiana Tech University. He researches interest rate determination and the inflationary consequences of suboptimal monetary policy. He teaches monetary economics, research methods & macroeconomic theory.