Federal Reserve Economic Data

St. Louis is the home of FRED and the FRED Blog and part of the Federal Reserve’s Eighth District. In the past few years, the housing markets in the seven states of our District have cooled and are slowly returning to their previous trend.

The FRED graph above plots the Federal Housing Finance Agency’s house price index for these states over the past five years: House price appreciation has moderated since its peak in 2022 and is now back to pre-pandemic levels, with prices increasing between 4% and 5% annually.

The second graph above shows the number of active house listings for each state relative to the number of listings in May 2017. Despite the normalization of price growth, the number of houses in most of these markets remains well below pre-pandemic levels. House inventories range from only 33% in Illinois to 99% in Tennessee, compared with the number in 2017.

It’s worth noting the positive relationship between the speed of house price appreciation and the increase in the supply of housing, seen most clearly here in Tennessee: Compared with the other states in our District, Tennessee had the fastest pace of price growth and also the fastest increase in the supply of housing.

How these graphs were created: First graph: Search FRED for “All Transactions House Price Index for Arkansas” and click on the first link. Click “Edit Graph” in the top right corner and change the units to “Percent Change from a Year Ago.” To quickly add the other states’ data, use this pattern of series IDs in the “Add Line” tab: “ARSTHPI” where the “AR” is for Arkansas, “ILSTHPI” where the “IL” is for Illinois, etc. Change the first date of the time frame to May 2017. Second graph: Search for “Housing Inventory: Active Listing Count in Arkansas” and click on the first link. Click “Edit Graph” to change the units to “Index (Scale Value to 100 for a chosen date),” choosing 2017-05-01. Add the series IDs in the same way as above, except the two letters that identify the state are at the end of the series IDs: “ACTLISCOUAR,” ACTLISCOUIL,” etc.

Suggested by John Fuller and Violeta Gutkowski.

The FRED Blog has discussed initial claims for state unemployment insurance benefits (initial claims, for short) as an economic indicator. Today, we discuss whether initial claims above 400,000 signal weakening labor market conditions.

Our FRED graph above shows weekly initial claims data reported by the US Employment and Training Administration from January 1967 to May 2025: The latest available value at the time of this writing is 229,000 on May 10. The data are displayed in a logarithmic scale to accommodate the large COVID-19-related spike of 6,137,000.

The dashed green horizontal line in the graph is positioned at a value of 400,000. This “rule of thumb” value is commonly referenced by business economists as the threshold, if crossed, that signals weakening labor market conditions. Recent research by Michael McCracken and Trần Khánh Ngân from the St. Louis Fed investigates the accuracy of this figure.

Their results show that the threshold value that is most accurate for signaling economic turning points varies significantly over time. And, since 1984, it stands around 434,165, which is significantly higher than in previous decades. Their results also show the optimal threshold value seems more informative for gauging labor market conditions during expansionary rather than recessionary periods.

For more about this and other research, visit the publications page of the St. Louis Fed’s website, which offers an array of economic analysis and expertise provided by our staff.

How this graph was created: Search FRED for and select “Initial Claims.” From the “Edit Graph” panel, use the “Add Line” tab to select the option “Create user-defined line.” Type “400000” in the boxes labeled “Value start/end.” Use the “Format” tab to select “Display > Log scale left.”

Suggested by Diego Mendez-Carbajo.

It can be useful to compare economic statistics across countries, including unemployment rates. It’s not always easy to do so, and our post today explains why.

The unemployment rate is a ratio of two measures:

1. the number of unemployed
2. the total labor force

Both measures are subject to their own definitions and interpretations, with some tricky details. For example, what counts as “looking for work”? This question defines who’s unemployed vs. who’s outside the labor force. Does perusing a job-posting website suffice, or does it require actively sending resumes to businesses? Do teenagers count? Students? What about temporary layoffs?

The Organisation for Economic Co-operation and Development (OECD) standardizes economic measures across its member countries and publishes harmonized statistics, including unemployment rates.

Our FRED graph above shows the OECD’s harmonized unemployment measure (in blue) along with the Bureau of Labor Statistics’ measure (in red) for the US. Both these rates use the age range of 25 to 54 years and track each other remarkably well, hinting that their definitions are very similar.

The US unemployment rate that’s widely disseminated, though, covers those who are 16 years or older. This rate (in green) is higher and should not be used to compare the US unemployment rate with the rates in other countries.

How this graph was created: Search FRED for and select “OECD unemployment rate US 25-54.” Click on “Edit Graph,” open the “Add Line” tab and search for “unemployment rate,” take the UNRATE series, and change the frequency to quarterly (average over months). Repeat for “unemployment rate 25-54.”

Suggested by Christian Zimmermann.