Skip to content
The Evergreen Edit
Save app to home screen
Business

How to Read the Monthly Retail Sales Report: What the Numbers Mean, With Worked Examples

A plain-English guide to the Census Bureau's monthly retail sales report: what it measures, why it isn't adjusted for inflation, how to tell if a change is real, and worked examples you can follow.

By Kiara Miller, Founder & Editor

Published · Last reviewed: · 9 min read

Checkout lanes with cash registers and bagging carousels inside a Walmart store
Photo: Mike Mozart via Wikimedia Commons (CC BY 2.0)

Retail sales headlines move markets and shape talk about the economy, but the report has a few quirks that change what the numbers really say.

What is the monthly retail sales report?

The monthly retail sales report is the U.S. Census Bureau's estimate of how many dollars Americans spent at stores, websites, car dealers, gas stations, restaurants and bars in the past month. It is not adjusted for inflation, it comes out about two weeks after the month ends, and its first estimate is revised later.

That's the short answer. The rest of this guide walks through what the report covers, the lines worth reading beyond the headline, and how to check whether a change is big enough to mean anything. It's written for anyone who sees "retail sales rose 0.4%" in the news and wants to know what that really says about shoppers, stores and the wider economy.

Why the report matters

The Census Bureau describes retail sales as one of the primary measures of consumer demand for both durable goods (things that last, like cars and furniture) and non-durable goods (things used up quickly, like food and gas). Because consumer spending is such a large part of the U.S. economy, economists, the Federal Reserve, investors and business owners all watch this report closely.

For a small business owner, it's a quick read on whether customers in your corner of retail are opening their wallets or pulling back. For everyday readers, it helps make sense of headlines about a "strong consumer" or "slowing spending." It sits alongside other reports we've explained, such as the Consumer Price Index and the monthly jobs report.

Where the numbers come from

The headline number comes from the Advance Monthly Retail Trade Survey. According to the Census Bureau, the advance estimates are based on a subsample of the Census Bureau's full retail and food services sample. About 4,800 firms are selected by stratified random sampling, and their answers are weighted and benchmarked to represent the universe of over three million retail and food services firms.

The survey covers retail trade, which is sector 44-45 in the North American Industry Classification System (NAICS), plus food services and drinking places (NAICS 722). That second group covers businesses that prepare meals, snacks and drinks to order, so restaurant and bar spending is part of the headline total.

Companies report the dollar value of their sales. A few details are easy to miss. The Census Bureau's FAQ says sales taxes are excluded from reported sales, while excise taxes are included. Gift card sales count when the card is redeemed, not when it's bought. Online shopping is counted too: the FAQ says e-commerce sales are included in the monthly retail sales estimates, and the Census Bureau also estimates them separately.

Timing matters as well. The Census Bureau says advance reports are released about 9 working days after the close of the reference month, usually at 8:30 a.m. Eastern time. So the report that comes out in the middle of a month describes the month before. The Census Bureau's economic indicator calendar lists the exact dates.

Quirk 1: The numbers are not adjusted for inflation

This is the most important point in the whole report. The Census Bureau says its estimates are adjusted for seasonal variation and holiday and trading-day differences, but not for price changes.

That means a rise in retail sales can come from people buying more things, from the same things costing more, or from a mix of both. If prices rose 3% over a year and sales rose 3%, shoppers didn't necessarily buy any more stuff. They just paid more for it. To get a rough sense of "real" spending, you compare the sales change with inflation, which is what the Consumer Price Index measures.

Quirk 2: The numbers are seasonally adjusted

Spending always jumps in December and slows in January. If the report didn't account for that, every December would look like a boom and every January like a bust. Seasonal adjustment removes those regular calendar patterns so one month can be compared fairly with the month before. The adjustment also accounts for holiday and trading-day differences, such as a month having one more Saturday than usual.

The Census Bureau also publishes unadjusted figures. Those are useful for comparing the same month across years, but headline month-to-month changes use the adjusted numbers.

Quirk 3: The first estimate is an early look and gets revised

Because the advance estimate is based on a subsample, it's revised as more data arrive. The Census Bureau says the average absolute revision to the advance estimate is two-tenths of one percent. That sounds small, but many monthly changes are also small, so a revision can turn a modest gain into a flat month, or the reverse. Each report shows a revised figure for the previous month, so it's worth reading that line too.

How to tell if a change is real

Every headline change in the report comes with a margin of sampling error, shown as "plus or minus" a number. The Census Bureau explains that this margin gives a range around the estimate that is a 90 percent confidence interval.

The rule for reading it is simple. If that range includes zero, the report says one does not have sufficient evidence to conclude at the 90 percent confidence level that the change is different from zero, and so the change is not statistically significant. The Census Bureau also notes that estimated changes shown in the text of the report are statistically significant unless otherwise noted.

In plain English: a small rise with a big margin of error might really be no change at all. Look for the plus-or-minus number before reading much into a headline.

The lines worth reading beyond the headline

The report's first table shows the total and several versions of it with certain categories removed. Each one answers a different question.

  • Retail and food services, total: the headline number, including restaurants and bars.
  • Total excluding motor vehicles and parts: removes car dealers and auto parts stores, whose big-ticket sales can swing the total.
  • Total excluding gasoline stations: removes gas stations, whose sales rise and fall with pump prices.
  • Total excluding motor vehicles and parts and gasoline stations: removes both, for a steadier view of everyday spending.
  • Retail: the total without food services, so you can see goods spending on its own.

The table also lists categories such as furniture stores, electronics stores, building materials dealers, grocery stores, health and personal care stores, clothing stores, general merchandise stores and nonstore retailers. If you run a business, the category closest to yours is often more useful than the headline.

Worked examples

These examples use made-up but realistic numbers so the math is easy to follow. Swap in the figures from an actual report to do the same checks yourself.

Example 1: Turning dollars into a percent change. Suppose total sales were $700.0 billion last month and $705.6 billion this month. The change is $705.6 billion − $700.0 billion = $5.6 billion. Divide by last month: $5.6 ÷ $700.0 = 0.008, or 0.8%. That matches the way the report states monthly changes.

Example 2: Checking whether the change is significant. Say a report shows sales up 0.3%, plus or minus 0.5%. The range runs from 0.3 − 0.5 = −0.2% to 0.3 + 0.5 = 0.8%. Because that range includes zero, the change isn't statistically significant at the 90 percent level. Now say sales rose 0.9%, plus or minus 0.4%. The range runs from 0.5% to 1.3%, which doesn't include zero, so the rise is significant.

Example 3: Adjusting for inflation. Say retail sales are up 4.0% from a year earlier, and the Consumer Price Index shows prices up 3.0% over the same year. A quick estimate of real growth is 4.0% − 3.0% = about 1.0%. The more precise way is to divide: 1.040 ÷ 1.030 = 1.0097, or about 0.97%. Either way, most of the headline gain came from higher prices, not from people buying a lot more. This is only a rough check, because the CPI covers more than retail goods, including rent and services.

Example 4: When gas prices drive the headline. Say total sales went from $700.0 billion to $704.2 billion, a rise of $4.2 ÷ $700.0 = 0.6%. Gas station sales went from $60.0 billion to $63.0 billion, up 5%, because pump prices jumped. Take gas stations out: the rest went from $640.0 billion to $641.2 billion. That's $1.2 ÷ $640.0 = about 0.19%. The headline looked solid, but most of the gain came from pricier fuel, and everyday spending outside the pump barely grew.

Example 5: Watching for revisions. Say last month's report showed sales up 0.5%. This month's report revises that month to up 0.7%. The earlier month was stronger than first thought, so the new month's change starts from a higher base. When you compare months, use the revised numbers, not the ones you remember from the news.

A step-by-step checklist for reading any retail sales report

When a new report comes out, these steps take about five minutes and give you a much clearer picture than the headline alone. Work through them in order.

  1. Find the month the report covers. It's the month before the release date.
  2. Read the headline monthly change and its plus-or-minus margin. Check whether the range includes zero.
  3. Read the revised figure for the previous month.
  4. Compare the total with the "excluding motor vehicles and parts and gasoline stations" line.
  5. Check the year-over-year change and compare it with the latest CPI to estimate real growth.
  6. Look at the category closest to your life or business.

Common mistakes to avoid

The biggest mistake is treating a rise in sales as proof that people are buying more. Because the figures aren't adjusted for price changes, inflation alone can push them up. Always look at prices alongside sales.

A second mistake is overreacting to one month. Small changes often fall inside the margin of error, and the first estimate will be revised. A trend over several months says more than any single report.

A third is ignoring the excluding lines. A spike in car sales or a jump in gas prices can make the headline look stronger or weaker than everyday spending really is.

A fourth is comparing unadjusted numbers month to month. Unadjusted December sales will almost always beat November's, because of the holidays. Use seasonally adjusted figures for month-to-month comparisons and unadjusted ones only for the same month across years.

Finally, comparisons between categories can be tricky. The report notes that comparisons across industries have not been tested for significance, so a bigger gain in one store type than another may not be a meaningful difference.

What to do next

Next time retail sales make the news, open the Census Bureau's release and run through the checklist above. If you also follow wholesale prices, our guide to the Producer Price Index shows how prices change before they reach store shelves. Reading the reports side by side gives you a fuller picture of whether spending is growing because people are buying more or because everything costs more.

Frequently asked questions

Are retail sales adjusted for inflation?

No. The Census Bureau says the estimates are adjusted for seasonal variation and holiday and trading-day differences, but not for price changes. To estimate real growth, compare the change with the Consumer Price Index.

When does the retail sales report come out?

The Census Bureau releases the advance report about 9 working days after the end of the month it covers, usually at 8:30 a.m. Eastern time. Exact dates are on its economic indicator calendar.

Do online sales count in retail sales?

Yes. The Census Bureau's FAQ says e-commerce sales are included in the monthly retail sales estimates. It also publishes separate e-commerce estimates.

Does buying a gift card count as a retail sale?

Not when it's bought. The Census Bureau's FAQ says gift card sales are counted when the card is redeemed.

Content on The Evergreen Edit is for educational and informational purposes only and is not individualized financial, investment, tax or legal advice. Learn more.

Which money-saving habit will you focus on next?

This article is for educational purposes only and isn't personalized financial, tax or legal advice. Consider talking with a qualified professional about your situation.

Kiara Miller

Kiara Miller — Founder & Editor

Kiara Miller is the founder and editor of The Evergreen Edit. He has been doing independent finance research for two decades, is a tax preparer, and reviews every article before it's published.

More about Kiara →

Get smarter about your money.

Join readers who get money stories on their chosen topics, daily or as new articles post.

Which topics do you want in your inbox? Pick up to 3
Email frequency

We'll use your answers to send articles you care about and to plan new content. We never sell your info. Privacy Policy

Recommended reading