Prices in 2026: What's Cheaper Than a Year Ago, and What Isn't

Prices overall were 3.4% higher in August 2026 than in August 2025, according to the Bureau of Labor Statistics (BLS). But that single number hides a split. Used cars and trucks were 2.3% cheaper than a year earlier, furniture and bedding were 0.8% cheaper and televisions were 1.0% cheaper. Clothing, on the other hand, was 3.6% more expensive, and energy was up 16.3%.

The headline inflation number is an average, and when you shop one aisle at a time, the average is rarely the price you pay.

Here is a clear look at the BLS figures, the date they cover, and what they suggest about where a careful shopper can find relief. All numbers below are the 12-month change in the Consumer Price Index for All Urban Consumers (CPI-U), U.S. city average, not seasonally adjusted. The BLS published them on 11 September 2026.

The table: what moved in the last 12 months

A few lines are worth a second look.

Where prices fell

Used cars and trucks are 2.3% cheaper than a year ago. That is one of the biggest drops in the table. Furniture and bedding are down 0.8%, and televisions are down 1.0%. These are big-ticket items people tend to buy rarely, so even a small drop is worth noticing.

A simple read: for things with a long life that you use at home, the price trend is not working against you right now. That does not mean every individual item is a deal. It means that, on average, you are not chasing a rising market.

Where prices rose faster than the average

Energy is the outlier at +16.3%, far above the +3.4% for all items. The BLS summary does not say here which kinds of energy drove it, so I will not guess. What it means for a household is that running costs matter more than they did a year ago. That has a practical edge for shoppers of used appliances, which we cover below.

Apparel is up 3.6%, which is above the overall rate. If you need clothes, this is the category where buying secondhand or from a clearance channel has the clearest advantage over paying retail price.

What this means for buying used and returned goods

The CPI measures what stores charge for new goods, so it is a benchmark, not a verdict on any particular listing. Here are some ways to use it:

1. Compare, don't assume. If used cars are 2.3% cheaper than a year ago, a seller who is holding firm on last year's price may be overestimating the market.
2. Think about running costs. With energy up 16.3%, an old appliance that wastes power can cost more than its low price suggests. Our guide to old refrigerators and the real cost of running them shows how to run that sum.
3. Time big purchases. Appliances are up only 0.1% and furniture is down, so you have less pressure to buy something just because you fear it will cost more next month. I cannot predict future prices, and neither can this table.
4. Set a ceiling in advance when bidding. The set your number first method works best when you know what the item costs new.

For shoppers in the DC, Maryland and Virginia area, the BLS also publishes regional numbers, which can differ from the national figure. I have not opened those pages, so I am only pointing out that they exist.

A word on what this data cannot tell you

The CPI tracks new goods and services at retail. It does not follow prices at auctions, in returned-goods channels or at private sales. So a 0.8% fall in furniture says nothing about the price of the exact dresser you are eyeing. Use it as a mood indicator for the market, then look at the condition, photos and terms of the actual item.

If you want to see what is available now, browse Buy Now or the live auctions, and read the help page for how bidding and pickup work.

The bottom line

Overall prices rose 3.4% in the year to August 2026, but used cars, furniture and TVs got cheaper, while clothing and energy rose faster than average. The practical takeaway is to compare prices category by category and to think about running costs, not just the sticker.

If you know someone who is shopping for a big item this fall, send them this table first.