May 2026

Turning thrifty: incorporating secondhand apparel into the Consumer Price Index

The secondhand clothing market has become a booming industry driven by consumers’ economic and sustainability concerns. This article discusses why and how the Consumer Price Index (CPI) program prices secondhand apparel by exploring the evolution of the secondhand clothing market. We also consider the improvements to item quality control that allow the CPI to incorporate secondhand garments, the weights used by the CPI to accurately reflect secondhand apparel expenditures, and the growth of nonclothing secondhand goods in other markets.

The Consumer Price Index (CPI) is the most widely used measure of inflation in the United States, measuring the average monthly change in prices paid by urban American consumers. It historically has limited its observed market basket of goods and services to new or unused items in good condition except for used cars and trucks. This restriction is to maintain a market basket with consistent item quality. However, given consumer preferences shifting toward secondhand goods and the expanding availability and improving quality control of secondhand apparel, the U.S. Bureau of Labor Statistics (BLS) incorporated secondhand apparel into the CPI in early 2025 and is further researching including other secondhand goods.¹

State of the wardrobe: market research and insights

The secondhand apparel market has had a long and varied history in the United States. To better represent current U.S. consumers' purchasing behaviors, BLS studied the market both as a whole and within the CPI sample.

Check out grandpa’s closet: the rise of thrifting

Retail secondhand clothing began to take shape in the late 19th century and early 20th century following the Industrial Revolution and the resulting higher turnover of clothing fashion.² Thrifting became more popular during the economic hardship of the Great Depression in the 1930s and during World War II in the 1940s, potentially due to wartime rationing.³ In the 1950s, Americans saw the emergence of high-end consignment stores and an increase in the popularity of garage sales.⁴ Since then, brick-and-mortar, donation-based thrift stores have been the largest used-clothing stores in the United States, although the amount of online secondhand stores and sales are rapidly growing as well.

Check out my closet: secondhand apparel expenditure since 1990

American households spent roughly the same amount of money on clothing each year between 1990 and 2022, except in 2020 when they reduced their spending on apparel by 23.8 percent during the early stages of the COVID-19 pandemic according to the Consumer Expenditure (CE) Surveys. Although the amount Americans spent on clothing has been consistent, the amount compared with other expenditure categories has been decreasing. Apparel’s relative importance in the CPI declined from 5.3 percent in 1993 to 4.0 percent in 2003, 3.4 percent in 2013, and 2.5 percent in 2023. In addition, the sales of multiple major clothing and department stores have suffered, with a higher-than-average number of stores declaring bankruptcy.⁵

Rather than shopping at major clothing and department stores, American consumers have been buying more secondhand clothing.⁶ The shift toward secondhand apparel is driven by consumer concerns about clothing costs and environmental sustainability, particularly within younger generations.⁷ More millennials and members of Generation Z are including secondhand items among their apparel purchases, with 42 percent of each generation shopping secondhand clothes at some point in 2020.⁸ In fact, over 20 percent of each generation was buying secondhand clothes each month in 2024.⁹ Upcycling (the repurposing of old or discarded items or materials) has also become a popular trend with Generation Z on social media.¹⁰

Although secondhand clothing was slowly expanding before 2020, its growth accelerated during the COVID-19 pandemic.¹¹ Americans were cleaning out their closets and buying more used clothes because they were spending more time at home and experiencing economic strain and inflation.¹²

In addition, consumers began shopping more at specifically online secondhand stores following the start of the COVID-19 pandemic.¹³ In 2022, online sales made up about 14 percent of one thrift retailer’s revenue at $251.8 million, with the associated website processing over 14,400 transactions per day.¹⁴

Online resale stores have especially surged in the apparel market. Resale stores generally sell higher end goods compared with the goods of traditional thrift stores.¹⁵ Resale stores are also distinct from donation-based thrift stores in that the previous owner receives a portion of the proceeds from the sale. Online resale stores comprise a growing segment of the clothing industry, with three of the larger retailers generating a combined total revenue of $803.7 million in 2021.¹⁶

Check out the Bureau’s closet: secondhand apparel in the CPI sample

Although BLS previously did not price secondhand items beyond used cars and trucks, BLS surveys captured consumer expenditures at secondhand stores for weighting purposes. As a result, secondhand stores have been included in the pool of stores to then be selected for the CPI sample but were ineligible for pricing until 2025.

Secondhand apparel made up 0.8 percent of total apparel-expenditure data using BLS survey data from the fourth quarter of 2016 to the third quarter of 2019. Overall, respondents tended to shop at chain brick-and-mortar secondhand stores, though they did spend more on average at online secondhand stores. They also spent the most money on women’s clothing when shopping secondhand apparel, comprising 42.8 percent of all secondhand apparel expenditures. (See chart 1.)

Though prices from items sold in secondhand clothing stores were not collected for use in the CPI, these stores would have comprised 1.5 percent of women’s apparel in the CPI sample in 2021, compared with 0.0 percent in 2009. (See chart 2.)

When looking within segments of the market, respondents also turned to secondhand stores for children’s clothing, with 1.3 percent of expenditures on all boys’ clothing and 1.4 percent of expenditures on all girls’ clothing bought secondhand. (See chart 3.)

The CPI program has evolved how it measures the secondhand apparel market. Since 2019, BLS stopped collecting expenditure data via telephone and replaced that data collection with the more representative CE Surveys.¹⁷ Accurately capturing the increase in secondhand clothing spending would not have been possible by continuing to rely on telephone collection. With the transition to CE data, more secondhand clothing stores have entered the CPI’s outlet sampling frame.

Constructing the wardrobe: concepts and methods

BLS has identified secondhand apparel as a growing segment of the apparel market. As a result, BLS incorporated secondhand apparel to better represent current U.S. consumers’ buying habits. However, secondhand goods require unique consideration and treatment compared to their new or unused counterparts when being incorporated into the CPI.

Quality control and ensuring a comparable basket of goods

Historically, BLS has not collected prices of secondhand or used goods because of the concern over item quality. The purpose of the CPI is to measure the changes of prices within a representative market basket of goods and services over time. To have a meaningful market basket, the measured goods and services must be of comparable quality over time. For instance, measuring the difference in price between a silk blouse and a cotton T-shirt has little meaning because the silk material is more expensive to produce than the cotton, thus the price difference does not represent a true price change. BLS matched-model methodology for the CPI maintains item quality over time using checklists composed of price-determining specifications. For example, the women’s tops checklist accounts for the type of top priced (sweater, pullover shirt, vest, and so on), fiber content, brand, sleeve length, and size range, among other factors. The previous single exception to the exclusion of secondhand goods in the CPI sample was used cars and trucks. Used cars and trucks are a separate index from new cars and trucks and are made up of used cars and trucks of 2 through 7 years of age. Used vehicles are controlled for mileage and age, maintaining item quality over time within the CPI sample.¹⁸

Unlike used cars and trucks, the quality of other secondhand goods is not as easily quantifiable. Continued use of clothing gradually thins the fabric, frays hems, and may cause damage to the finer details like any decorative embroidery, sequins, or beading. Following our earlier example, a like-new, barely worn cotton T-shirt would have better quality and a resulting higher value than a threadbare cotton T-shirt.

However, the major secondhand clothing retailers have evolved over time to account for differing item quality, for both brick-and-mortar thrift stores and online resale stores alike. Major brick-and-mortar thrift stores accept and sell new and gently used donations.¹⁹ Online resale stores can provide more detailed descriptions of item quality. It is standard practice for secondhand item listings to include a glossary of terms for sellers to include in their listings, from “brand new” to “very used.” Many popular chain secondhand apparel stores have measures in place to control item quality. As a result, BLS will follow data collectors' observations and stores' descriptions to maintain consistent quality over time and will continue to exclude goods that are damaged.

Collecting and weighing the expenditures

The CPI uses two main components to measure price change: prices and weight. For example, let us say that the price of spacecraft equipment has increased 1000 percent. However, as there is only a handful of spacefaring private individuals in the United States, this will not cause a large change in the expenses of the average consumer. Conversely, like in early 2023 and 2025, the price of eggs rose noticeably.²⁰ Because many Americans regularly buy eggs or other goods that use eggs, this increase affected the average consumer and contributed to inflation.

Secondhand goods have unique issues related to accurately capturing expenditure information and the resulting weights assigned to them. If one household sold a pair of shoes to another household, the combined net change in cost of living is zero because they cancel each other out. Thus, household-to-household exchange is conceptually excluded from the CPI scope.²¹ For this reason, as well as operational logistics, yard sales and private sales are excluded from the CPI sample.

Thrift stores operate entirely from donations in which the donating household receives no money, so the price paid goes entirely to the store and is within scope of the CPI. For resale stores, the CPI would ideally weight secondhand goods by the commission earned from each secondhand good transaction. However, this information is often not readily available. In addition, in many resale instances, sellers can negotiate prices with consumers, making it difficult to estimate the commission before a transaction is complete.

An alternative weighting method for resale stores would be to ask consumers in the CE survey how much they earn by selling apparel, in addition to the typical questions about how much they spend. The CPI would then measure the total household revenue and total household expenditures on apparel, then take the difference. The difference would be used for weighting items priced at secondhand apparel stores. However, this would double the burden on respondents to the CE survey, leading to lower respondent rates and lower quality responses. Thus, BLS has decided to not weight secondhand apparel goods by net expenditure and will continue to weight them similarly to new or unused goods. BLS economists will revisit alternative data sources with this information in the future.

Other secondhand items

In addition to clothing, the market for secondhand goods has been expanding in many other industries. Americans also buy books and media, electronics, and furniture secondhand. Particularly, American consumers are purchasing more used textbooks and refurbished electronics.²² The U.S. Census Bureau estimates that used merchandise stores made $24.3 billion in sales in 2022, compared with $5.6 billion in 1992 and $14.7 billion in 2012. (See chart 4.)²³

However, these goods pose an additional hurdle to being included in the CPI because of their wider availability of collectible and antique goods. Collectible and antique goods are investment goods that appreciate in value over time, such as first edition books and antique furniture. Spending on investments is considered distinct from spending on consumption and thus is out of scope for the CPI under the cost-of-living framework that guides the index.²⁴ As such, CPI sample must continue to exclude collectible, vintage, and antique goods. BLS economists are researching how these secondhand goods would be incorporated into the CPI, if at all.

Rounding out the wardrobe

The reuse and recycling of clothing has expanded into a thriving and growing market in recent years because of consumer concerns over clothing costs and environmental sustainability, particularly following the COVID-19 pandemic. The secondhand apparel market has also evolved, improving quality control and allowing consistent item comparability from month to month.

As markets and consumers’ tastes evolve, so too must the CPI market basket. As of early 2025, BLS began including secondhand apparel in its pricing sample, achieving a long-standing goal to more accurately represent the ever-changing apparel market.