FRED is a free database where you can find official unemployment data going back decades

FRED stands for Federal Reserve Economic Data. It is a searchable online tool run by the Federal Reserve Bank of St. Louis that holds thousands of economic statistics, including every unemployment rate the U.S. government has published since 1948. You do not need an account to use it. You search for a data series by name or code, and FRED shows you the numbers in a table, a graph, or a downloadable file.

The unemployment rates you find there come from the Bureau of Labor Statistics (BLS), a division of the U.S. Department of Labor. FRED does not calculate the rates itself—it republishes official government data and lets you organize it by date, geography, or demographic group. If you are reading about unemployment in a news article or a policy paper and you want to see the actual numbers behind the claim, FRED is where you go to check.

Most people use FRED for one of three reasons: to see how unemployment has moved over time, to compare rates across states or cities, or to read data for their own analysis. The tool is built for people who want to dig into the numbers themselves rather than rely on a summary someone else wrote.

Key Takeaways

  • FRED publishes the same unemployment data the BLS releases each month, so the numbers are official and current.
  • You can search by series name (like "unemployment rate" or "civilian labor force") or by series ID code, which is faster once you know what you are looking for.
  • FRED lets you graph data across any time period, compare multiple series side by side, and read the results as a spreadsheet or image.
  • The unemployment rate FRED shows you is the seasonally adjusted rate for the total civilian labor force unless you specify otherwise.

How to find unemployment data on FRED

Go to fred.stlouisfed.org and use the search box at the top. Type "unemployment rate" and you will see a list of series. The first result is usually "Unemployment Rate" (series ID UNRATE), which is the headline number you hear in the news—the percentage of the civilian labor force that is jobless and actively looking for work, seasonally adjusted, for the entire United States.

If you want a different slice of the data, the search results show you other options. You can find unemployment by state (search "unemployment rate [state name]"), by age group (search "unemployment rate 16-19" or "unemployment rate 25-54"), by race or ethnicity, or by education level. Each series has its own ID code. Once you find the series you want, click on it to open the data page.

On the data page, you see a graph and a table. The graph shows the trend over time. The table lists the exact number for each month or quarter, depending on the series. You can change the date range by clicking on the graph or typing dates into the boxes at the top. You can also read the data as a CSV file (for spreadsheets) or an image (for presentations or reports).

What the numbers mean and why they matter

The unemployment rate is a percentage, not a count. It answers the question: of all the people in the labor force, what share is currently unemployed? The labor force includes people who are working or actively looking for work. It does not include people who are retired, in school full-time, disabled and not seeking work, or have stopped looking for a job. This matters because the unemployment rate can stay the same or even fall while the number of jobless people rises, if enough people leave the labor force.

FRED shows you the seasonally adjusted rate by default. Seasonality means that some industries hire more people at certain times of year—retail hires for the holidays, construction slows in winter, schools hire in summer. The BLS adjusts the raw numbers to remove these predictable swings so you can see the underlying trend. If you want the unadjusted rate, FRED has that too (search "unemployment rate, not seasonally adjusted").

The unemployment rate is one measure of labor market health, but it is not the only one. FRED also has data on the labor force participation rate (the share of the population that is working or looking for work), the employment-population ratio, and the number of people who have stopped looking for work. Comparing these numbers together gives you a fuller picture than the unemployment rate alone.

Comparing unemployment across states and time periods

One of FRED's most useful features is the ability to graph multiple series at once. You can see how unemployment in your state compares to the national average, or how it has changed over the past five years. To do this, search for the national rate (UNRATE), open it, then use the "Add Data Series" button to add your state's rate. FRED will plot both lines on the same graph so you can see them side by side.

State unemployment rates are published monthly by the BLS, usually with a one-week lag after the national rate. Some states have higher baseline unemployment than others because of industry mix, population demographics, or economic structure. Comparing your state's rate to the national average tells you whether your state is doing better or worse than the country as a whole, but it does not tell you whether the rate is "good" or "bad" in absolute terms—that depends on context and what economists expect.

You can also use FRED to look at unemployment during specific events or time periods. Search for "unemployment rate" and graph it from 2007 to 2009 to see the Great Recession, or from 2020 to 2021 to see the COVID-19 shock and recovery. The visual comparison often makes patterns clearer than reading numbers in a table.

Understanding FRED's data sources and limitations

All unemployment data on FRED comes from the BLS Current Population Survey (CPS), a monthly survey of about 60,000 households. The BLS asks whether people in those households are working, looking for work, or neither. From those responses, it calculates the unemployment rate for the nation and for subgroups. FRED republishes these official numbers without changing them.

The CPS is the most reliable source of unemployment data the U.S. government produces, but it is still a survey, not a count of every person. This means there is always some margin of error, especially for small subgroups or small geographic areas. FRED does not display confidence intervals or margins of error on its graphs, so if you are using the data for research or policy work, you should check the BLS website directly for the technical details.

FRED updates unemployment data monthly, usually on the first Friday of the month when the BLS releases the jobs report. The data is never revised after the initial release for that month, though the BLS does revise the prior two months' figures. If you are tracking a specific month's number, check whether FRED is showing you the original release or a later revision.

Using FRED data for your own analysis

Once you read data from FRED, you can use it in a spreadsheet, a statistical program, or a visualization tool. The CSV read includes the date and the value for each observation, which you can paste directly into Excel or Google Sheets. If you want to create your own graph or calculate changes over time, this is the easiest way to do it.

FRED also has an API (process programming interface) for people who write code. If you use Python, R, or another programming language, you can write a script that pulls data from FRED automatically. This is useful if you want to update your analysis every month without manually downloading new files. The FRED website has documentation on how to use the API.

A common use of FRED unemployment data is to calculate the change in the unemployment rate over a period—for example, "unemployment fell 0.5 percentage points in the past year." Be careful with language here: a change from 5% to 4.5% is a 0.5 percentage point drop, not a 10% drop. The difference matters when you are communicating the numbers to others.

When FRED unemployment data does not tell the whole story

The unemployment rate measures joblessness, but it does not measure underemployment (people working part-time who want full-time work), wage stagnation, or job quality. Someone who is working one hour per week is counted as employed. Someone who has been out of work so long they stopped looking is not counted as unemployed. FRED has data on these other measures too (search "underemployment rate" or "labor force participation rate"), but you have to look for them deliberately.

Unemployment also varies widely by education, age, race, and geography. The national rate can mask very different conditions in different groups. If you are trying to understand labor market conditions for a specific population, search FRED for that demographic breakdown rather than relying on the headline number.

Finally, the unemployment rate is a backward-looking measure. It tells you what happened last month, not what will happen next month. If you are trying to predict whether unemployment will rise or fall, you need to look at other indicators like job openings, initial jobless claims, or business confidence surveys. FRED has those too.

Frequently Asked Questions

Can I see unemployment data for cities or counties on FRED?

FRED has state-level unemployment data, but not city or county data. For local unemployment rates, you need to go to the BLS website directly and use their Local Area Unemployment Statistics (LAUS) tool. Some states also publish their own local unemployment data through their labor departments.

Why does the unemployment rate sometimes go down when jobs are lost?

This happens when people leave the labor force faster than they lose jobs. If 100,000 people stop looking for work and 50,000 lose their jobs, the unemployment rate can fall even though the total number of employed people dropped. FRED has labor force participation data so you can check whether this is happening.

How far back does FRED unemployment data go?

The national unemployment rate on FRED goes back to January 1948. Some demographic breakdowns (by age, race, or education) have shorter histories because the BLS did not collect that data consistently until later. The data page for each series shows you the start date.

Is FRED data the same as what I see in news articles about unemployment?

Yes. News articles about the monthly jobs report cite the same BLS data that FRED publishes. FRED just makes it easier to search, graph, and read the numbers yourself rather than reading someone else's summary of them.

Can I use FRED data for a school project or research paper?

Yes. FRED data is public and free to use. If you read data and use it in a paper, cite the source as the Federal Reserve Economic Data (FRED) and note the series ID and the date you accessed it. This makes it straightforward for someone else to find the same data later.