The FRED® Blog

Squinting at consumer sentiment: Recent European pessimism

The FRED graph above is data heresy: It jams 12 time series into one graph. Why would we do this? Actually, we wanted more than 12 series, but even FRED graphs have their limits.

We suppressed the titles of the series so we really do need to explain what we’re doing here. 1. Each line represents a European country. 2. The data are about consumer sentiment. And 3. The idea is to show general patterns across Europe, without concern for any particular country. Basically, we’re looking for a flow or trend to emerge, using only the squinting-eye technique we discussed in our previous blog post. And we do see two distinct dips common to all the lines!

The OECD surveyed people about their opinion on where their economy is heading: Lower scores show pessimism, and higher scores show optimism. Europeans’ dispositions are spread out, but over the past 5 years they show an overall tendency toward pessimism about their economies. And there are two instances of shared pessimism across Europe: April 2020 and March 2022. The first date is when the COVID-19 pandemic started to shut down entire economies. The second date is the start of the invasion of Ukraine. Clearly, non-economic events can have a large and quite sudden economic impact.

Which countries are shown? Click on the graph to see the series titles, which include the country names: Austria, Belgium, Czech Republic, Denmark, France, Finland, Greece, Hungary, Ireland, Netherlands, Poland, and Switzerland. Btw, The two apparent outliers are Greece (in burgundy) and Switzerland (in turquoise).

How this graph was created: Search FRED for “OECD consumer sentiment.” Check all the countries you want displayed, up to 12. Click on “Add to Graph.” Click on “Edit Graph,” open the “Format” tab, and uncheck the titles display. Restrict the sample period to the past 5 years.

Suggested by Christian Zimmermann.

Unemployment by state: The lows in Dec 2022 vs. the peak in April 2020

The US economy in general has gone on a rollercoaster ride since the onset of the pandemic. Unemployment is one specific bumpy example: In April 2020, the unemployment rate was higher than it had ever been, but by December 2022 it was close to a record low.

We could ask many related questions here, but today we look at the unemployment rates across US states during this extreme episode.

The FRED map above shows the unemployment rate for each state in December 2022, the last date available at the time of this writing. We see quite a bit of variation, from 2.2% in Utah to 5.2% in neighboring Nevada. To look for big patterns, we can use the squinting-eye technique (patent pending!) to see the general distribution of color. In this map, the unemployment rate is higher (darker) in the Rust Belt and the West and lower (lighter) in the central and Mountain areas.

The second map also shows the unemployment rate for each state, but back in April 2020, when it was much higher everywhere. In fact, the lowest rate in April 2020 (Wyoming, 5.4%) was higher than the highest rate in December 2022 (Nevada, 5.2% ). By the way, in April 2020, the highest rate is again in Nevada (28.5%).

And what does the squinting-eye technique show us here? Again, the unemployment rate is higher (darker) in the Rust Belt and the West and lower (lighter) in the central and Mountain areas. In other words, at least in this particular economic episode, the whole nation seems to have gone up and down while maintaining its broad state-to-state differences.

How these maps were created: Search FRED for the unemployment rate. Click on any state rate, then click on “View Map.” To have the color span the unemployment rates over the two very distinct time periods, change the date to 2020-04-01, click on “Edit Map,” change the number of color groups to four, write down the interval values, and repeat for 2022-12-01. Now change the number of color groups to eight and select data grouped by “user defined method.” Enter all eight interval values. You have the first map. Change the date to 2020-04-01 and you have the second map.

Suggested by Christian Zimmermann.

Was there a tech-hiring bubble? job postings data suggest so

We recently discussed Information industry jobs data from the U.S. Bureau of Labor Statistics: That post showed that layoffs in Information were elevated relative to layoffs for non-farm employment overall. Today we examine job postings data from to answer a related question: How have the help wanted ads in the US tech industry compared with the rest of the labor market?

The FRED graph above taps into a recently expanded dataset from to compare changes in the level of job postings for the tech industry with changes in the level for all job postings. The black dashed line shows the indexed trend of all job postings; the red line shows the trend of postings in information technology operations and helpdesk positions; and the blue line shows the trend in software development job postings. These daily data are reported as an index with a value of 100 on February 1, 2020, and represent changes in the level of postings relative to that date.

The graph shows that the overall level of job postings declined during the onset of the COVID-19 pandemic, bouncing back to pre-pandemic levels 12 months later. Soon after, and while help wanted ads for IT operations and helpdesk staffing remained depressed, the level of job postings in software development started to quickly outpace the rising national trend.

At their peak in late February 2022, help wanted ads for designing computer applications or programs more than doubled their February 2020 level. In contrast, IT operations and helpdesk staffing roughly kept up with the national trend. Over the past year, the decline in software development job postings has been swift and, again, has outpaced the gradual decline in overall job postings.

Finally, other datasets from the same source show similar ups and downs in other countries around the world (see here and here). So, although the data from is not as comprehensive as the BLS data, they suggest a large-scale cycle in tech industry employment is over.

How this graph was created: Search FRED for “Software Development Job Postings on Indeed in the United States.” Next, click the “Edit Graph” button, select “Add Line,” and search for “IT Operations and Helpdesk Job Postings on Indeed in the United States.” Repeat the last step to add “Job Postings on Indeed in the United States.” Use the “Format” tab to change the lines style and color.

Suggested by Diego Mendez-Carbajo.