How Americans Spend Their Time and Money: Three New Datasets on PUMSdata

PUMSdata started with one question: who lives where, and how? Today we are adding three datasets that answer a different set of questions, all from the U.S. Bureau of Labor Statistics. Put simply, they cover how Americans spend their time, what their work pays, and where their money goes. Each one is now live, in the same map, chart, and table tools you already know.

Here is the short version of what arrived:

• Time Use tells you how the average person divides up a 24-hour day, from sleep to work to leisure. See how Americans spend their day.
• Employment & Wages tells you how many jobs there are and what they pay, by industry, for every state, county, and metro area. Browse employment and wages.
• Spending tells you what the typical household spends in a year, category by category. See where household money goes.

Time Use: your 24 hours, on average

The American Time Use Survey (ATUS) asks thousands of Americans to walk through a single day in detail, hour by hour, and record exactly what they were doing. Add all those diaries up, apply the survey weights so the sample represents the whole country, and you get a picture of the average American day. It is one of the few places you can see, in real numbers, the trade-offs everyone feels: time spent working versus caring for family, or scrolling versus sleeping.

Here is the average American day in the most recent year of data, hour by hour:

Why does work show up as only about three hours? Because these are averages across everyone and every day, including weekends, retirees, children, and people who did not work that day. Spread the country's total working hours across all 365 days and all people, and the daily average comes out low even though a working weekday is long. That is exactly why we also show two other views for each activity: the share of people who did it at all, and the average time among just the people who did it. Sleep is nearly universal; competitive sports are not, so their all-person average is tiny while their among-doers average is not.

The real value is in the breakdowns. You can split any activity by sex, age, education, employment status, whether there are children in the household, and even weekday versus weekend, or define your own group. Compare how much time parents of young children spend on childcare versus adults without kids, or how leisure time shifts across age groups. Open the Time Use explorer and start pulling it apart.

Employment & Wages: what work pays, everywhere

The Quarterly Census of Employment and Wages (QCEW) is about as close to a full count of American jobs as exists. It is built from the payroll records that employers file for unemployment insurance, so it covers roughly 95 percent of all jobs in the country. This is not a small survey sample; it is the near-complete ledger. In the most recent quarter it counts about 157 million jobs paying an average of roughly $1,570 a week, and because it is geographic, you can put that on a map:

What makes it powerful is the detail. You can see, for any place, three core measures: how many jobs there are, what the average job pays per week, and how many separate employers (worksites) there are. And you can slice all of it by industry sector, from manufacturing and construction to health care, finance, and hospitality. That means you can answer very specific questions: what does a tech job pay in Austin versus Denver, which counties added the most health care jobs last year, or how manufacturing wages in one metro compare to the national average.

We have loaded 14 years of it, from 2012 through the latest 2025 quarter, at three levels of geography: every state, roughly 3,140 counties, and about 390 metro areas. Because the data is quarterly, you can watch a local job market move over time, not just check a single snapshot. Compare employment and wagesacross places, or open the map explorer and switch the source to Employment & Wages.

Spending: where the household budget goes

The Consumer Expenditure Survey (CE) is the government's accounting of how families actually spend their money. It is the survey behind a lot of familiar facts, like housing being the biggest line in most budgets, and it is the same data used to build the market basket for inflation. We have turned it into a plain-language picture of the average household's year.

In the most recent year, the average U.S. household spent about $73,800. Here is where it went:

Averages hide the interesting part, though, which is how spending changes with circumstance. You can break the budget down by income, age, region of the country, household size, whether the family owns or rents, and education. That is where the story lives: how much more of a lower-income household's budget goes to the basics, how transportation costs rise in some regions, or how a family's spending shifts as it grows. Explore household spending and see for yourself.

Why these three, together

Time, work, and money are the three things every household is constantly balancing, and until now the numbers behind them lived in three separate, hard-to-read government releases. Putting them in one place, in the same familiar interface, lets you connect them. A single question like “what does life cost and look like for a family in this part of the country” used to require three different tools and a lot of patience. Now it is a few clicks.

These datasets also round out the picture our other data already paints. Our Census PUMS data tells you who lives somewhere and the shape of their households; our building permits datatells you what is being built there next. Now Time Use, Employment & Wages, and Spending fill in the daily rhythm, the local economy, and the household budget.

One thing worth understanding

If you are used to our PUMS tools, these three datasets work a little differently, and the difference is worth a sentence. PUMS is microdata: a sample of individual records you tabulate yourself. These three are published estimates. The Bureau of Labor Statistics has already done the weighting and the math and released the finished numbers, so you are reading and comparing official figures rather than building them from raw records. That makes them fast and simple to work with. Employment & Wages is a near-complete count; Time Use and Spending are survey-based estimates, which means the national and large-group figures are solid, while very small subgroups should be read as approximate.

A few honest notes

A little context helps you read these numbers well. Time Use and Spending are surveys, so every figure carries some sampling uncertainty, and the smaller the group you slice down to, the noisier the estimate. Time Use is national only, with no state or local geography, because the survey is not large enough to support it reliably. Spending reflects what households report, and our figures track the Bureau's published totals closely, reading a few percent lower in some categories by design because of how we rebuild them from the underlying detail; the methodology pageexplains that in full. Employment & Wages counts jobs, not people, so someone with two jobs is counted twice, and very small industry cells in a given county are sometimes withheld by the Bureau to protect individual employers. As always, the data comes from the U.S. Bureau of Labor Statistics; PUMSdata is an independent product and not affiliated with the Bureau.

Start exploring

All three are live right now. Pick the question that interests you and jump in: how Americans spend their day, what work pays across the country, or where the household budget goes. If a public dataset you rely on is not here yet, tell us; the list keeps growing.