Community supported agriculture statistics at a glance
Community supported agriculture has moved from a niche farm model into a measurable part of direct food sales. The numbers below show how CSA farms, membership structures, and marketing channels have shifted across multiple USDA and CIAS sources.
Table of contents
- At-a-glance numbers
- The long-run CSA timeline
- What the USDA local-food counts show
- Who CSA operators are
- Farm size, land use, and production style
- Membership structure and season length
- Where CSA farms sell beyond the box
- What demand and share data say
- What the survey signals about growth and difficulty
At-a-glance numbers
A few figures capture the scale of the category quickly.
- USDA NASS reported 12,617 farms marketed products through a CSA arrangement in the 2012 Census of Agriculture (USDA NAL CSA page).
- USDA NASS reported 12,549 farms marketed products through a CSA arrangement in 2007 (USDA NAL CSA page).
- USDA NASS reported 7,244 farms sold through a CSA arrangement in 2020 (USDA NAL CSA page).
- CSA sales were $226 million in the 2015 USDA local food survey (USDA NAL CSA page).
- CSA sales were $225 million in the 2020 USDA local food survey (USDA NAL CSA page).
- The average share of farm sales coming from the CSA was 53.2 percent in the USDA AMS 2014 CSA manager survey (USDA AMS 2017 CSA report).
- For 58.1 percent of surveyed CSAs, the CSA was the major market channel (USDA AMS 2017 CSA report).
- The 2022 direct-sales total was $17.5 billion in direct food sales through direct marketing channels (USDA NAL CSA page).
Quick read on the scale
The clearest pattern is that CSA is not just a consumer buying club. It is a commercial marketing channel that can account for most of a farm’s sales, and in the USDA AMS survey more than half of respondents said CSA was their major market channel (USDA AMS 2017 CSA report).
At the same time, the national direct marketing market is much larger than CSA alone. The 2022 Census of Agriculture found $17.5 billion in direct food sales through direct marketing channels, which sets CSA inside a broader direct-to-consumer economy rather than treating it as a stand-alone niche (USDA NAL CSA page).
The long-run CSA timeline
The history data in the source material shows a slow build, then sustained maturity.
- The U.S. CSA concept began in Massachusetts in 1986 (University of Kentucky CSA page).
- By 1990, there were 60 CSA farms in the United States (University of Kentucky CSA page).
- The 1999 national CSA survey database contained 825 CSA farmers (CIAS 2004 Research Brief #67).
- The 1999 national CSA survey had 316 farm responses from 41 states (CIAS 2004 Research Brief #67).
- About 65 Wisconsin community supported farms were expected to grow food for an estimated 3,000 households in 1996 (CIAS 1996 Research Brief #21).
The early counts do not imply a smooth national rollout. They show a model that spread from one state-level origin into a wider farm strategy over time, with enough variation by region and operation type that later surveys could capture substantial differences in farm size, management, and sales structure.
What the USDA local-food counts show
The USDA numbers on CSA farms and sales show a category that has remained sizable even as the surrounding direct-market economy changed.
| Year | CSA farms or operations | CSA sales | Context |
|---|---|---|---|
| 2007 | 12,549 farms marketed products through a CSA arrangement (USDA NAL CSA page) | Not stated | Baseline census figure |
| 2012 | 12,617 farms marketed products through a CSA arrangement (USDA NAL CSA page) | Not stated | Slightly above 2007 |
| 2015 | 7,398 farms sold directly to consumers through a CSA arrangement (USDA NAL CSA page) | $226 million (USDA NAL CSA page) | Local food survey year |
| 2020 | 7,244 farms sold through a CSA arrangement (USDA NAL CSA page) | $225 million (USDA NAL CSA page) | Local food survey year |
| 2022 | 116,617 farm operations selling directly to consumers (USDA NAL CSA page) | $17.5 billion in direct food sales (USDA NAL CSA page) | Direct-marketing total |
The table is not a single apples-to-apples series, but the direction is still useful. CSA-specific counts are well into the thousands, while the broader direct marketing universe is much larger. That separation matters when you interpret the category: CSA is meaningful, but it is one channel inside a larger direct-sales system.
The census-to-survey shift
One useful comparison is the movement from 2007 to 2012. USDA NASS reported 12,549 farms in 2007 and 12,617 in 2012, and the 2012-to-2007 change in CSA farms was 0.5 percent (USDA NAL CSA page).
That near-flat change suggests stability rather than explosive expansion. It also implies that the category had already become established enough by the 2007-2012 period to hold a broad base of farms, even as the direct food market kept evolving.
Sales stayed near the same level across two survey years
The sales figures are also notable for their consistency. CSA sales were $226 million in the 2015 USDA local food survey, and CSA sales were $225 million in the 2020 USDA local food survey (USDA NAL CSA page).
That is not a dramatic jump, but it does show persistence. The share figure gives it more context: CSA sales were about 7 percent of the $3 billion in 2015 direct-to-consumer sales, and CSA sales were about 7.75 percent of the $2.9 billion in 2020 direct-to-consumer sales (USDA NAL CSA page).
Who CSA operators are
The 1999 and USDA AMS survey figures show a group that is comparatively experienced, often younger than the broader farm population, and relatively diverse in gender representation.
- The average age of primary CSA farmers was 43.7 years (CIAS 2004 Research Brief #67).
- Primary CSA farmers were 10 years younger than the average U.S. farmer in the 1997 Census of Agriculture (CIAS 2004 Research Brief #67).
- Forty-three percent of primary CSA operators had fewer than 10 years of farming experience (CIAS 2004 Research Brief #67).
- Primary CSA operators averaged 5.4 years as CSA farmers (CIAS 2004 Research Brief #67).
- About 39 percent of primary CSA operators were female (CIAS 2004 Research Brief #67).
- Only 8.6 percent of primary operators in the 1997 Census of Agriculture were female (CIAS 2004 Research Brief #67).
The 2014 USDA AMS manager survey adds another layer
The manager survey provides a later snapshot of who was running CSA operations.
- In the USDA AMS 2014 CSA manager survey, 259 respondents were female (USDA AMS 2017 CSA report).
- In the USDA AMS 2014 CSA manager survey, 171 respondents were male (USDA AMS 2017 CSA report).
- The USDA AMS survey recorded 7 respondents age 18-24, 102 respondents age 25-34, 116 respondents age 35-44, 92 respondents age 45-54, 87 respondents age 55-64, and 26 respondents age 65+ (USDA AMS 2017 CSA report).
A practical takeaway from those counts is that the category was concentrated in working-age operators, especially in the 25-44 range, with fewer very young and very old respondents. The exact distribution matters because CSA management tends to require hands-on coordination across harvest, packing, member communication, and delivery.
Farm size, land use, and production style
The older survey data show that many CSA operations were relatively small in land base and often leaned heavily into organic or biodynamic methods.
- Twenty-seven percent of surveyed CSA farmers owned zero acres (CIAS 2004 Research Brief #67).
- Another 17 percent owned 10 or fewer acres (CIAS 2004 Research Brief #67).
- Seventy-seven percent of farms had fewer than 30 acres of cropland (CIAS 2004 Research Brief #67).
- Nearly 58 percent used at least half of their land for the CSA operation (CIAS 2004 Research Brief #67).
- Thirty-seven percent used more than 90 percent of their land for CSA (CIAS 2004 Research Brief #67).
- Forty-one percent of surveyed CSA farms were certified organic (CIAS 2004 Research Brief #67).
- Ninety-four percent used organic or biodynamic practices (CIAS 2004 Research Brief #67).
What those land figures mean
The land-use data point to farms that were often highly committed to the CSA model rather than using it as a side channel. If 37 percent used more than 90 percent of their land for CSA and nearly 58 percent used at least half their land for CSA, then the model was frequently central to the farm’s identity and revenue strategy, not just an add-on.
The organic figures reinforce that interpretation. Certification was common, but practice-based alignment was even broader: 94 percent used organic or biodynamic practices, while 41 percent were certified organic (CIAS 2004 Research Brief #67).
Membership structure and season length
Membership design is one of the features that makes CSA distinct from other direct sales channels. The available data show a relatively structured membership model with room for variation.
- The median CSA farm sold 29 large memberships (CIAS 2004 Research Brief #67).
- The median CSA farm sold 23 smaller memberships (CIAS 2004 Research Brief #67).
- Ninety-four farms reported some kind of other membership such as home delivery, flower, winter, or institutional memberships (CIAS 2004 Research Brief #67).
- The average CSA delivery season lasted 23 weeks (CIAS 2004 Research Brief #67).
- Seventy-two percent of farms in the survey did not have a core group (CIAS 2004 Research Brief #67).
A short season, but a structured commitment
A 23-week average delivery season is long enough to require planning and customer retention, but short enough that farms must be precise about timing and consistency (CIAS 2004 Research Brief #67). The membership counts also suggest a business model that can scale by repeating a defined offer rather than by constantly redesigning the product.
The fact that many farms offered other membership types suggests flexibility. Home delivery, flower, winter, and institutional memberships show that CSA logic can extend beyond the classic weekly produce box while still staying inside the same direct-relationship framework (CIAS 2004 Research Brief #67).
Where CSA farms sell beyond the box
The USDA AMS survey makes clear that CSA farms often diversify their channels. The percentage figures show a layered sales strategy rather than a single-channel business.
| Sales channel | Share of CSA respondents |
|---|---|
| Farmers markets | 64.8 percent (USDA AMS 2017 CSA report) |
| Restaurants | 55.0 percent (USDA AMS 2017 CSA report) |
| On-farm retail | 41.4 percent (USDA AMS 2017 CSA report) |
| Grocery channels | 38.4 percent (USDA AMS 2017 CSA report) |
| Schools or institutions | 19.6 percent (USDA AMS 2017 CSA report) |
| Other channels | 16.2 percent (USDA AMS 2017 CSA report) |
| Contract to processors | 3.6 percent (USDA AMS 2017 CSA report) |
| Auction sales | 2.6 percent (USDA AMS 2017 CSA report) |
Location also shaped channel mix
The survey breaks out several channels by location type, and the pattern is not uniform.
- In the USDA AMS survey, 104 CSAs were located within 50 miles of a large city, 143 were located within 50 miles of a small city, 114 were in small towns, and 69 were in the countryside (USDA AMS 2017 CSA report).
- 60 percent of CSAs near large cities also used farmers markets, 69 percent of CSAs near small cities also used farmers markets, 69 percent of small-town CSAs also used farmers markets, and 68 percent of countryside CSAs also used farmers markets (USDA AMS 2017 CSA report).
- 47 percent of CSAs near large cities used on-farm retail, 37 percent of CSAs near small cities used on-farm retail, 35 percent of small-town CSAs used on-farm retail, and 56 percent of countryside CSAs used on-farm retail (USDA AMS 2017 CSA report).
- 32 percent of CSAs near large cities sold through grocery channels, 38 percent of CSAs near small cities sold through grocery channels, 44 percent of small-town CSAs sold through grocery channels, and 43 percent of countryside CSAs sold through grocery channels (USDA AMS 2017 CSA report).
- 58 percent of CSAs near large cities sold to restaurants, 61 percent of CSAs near small cities sold to restaurants, 54 percent of small-town CSAs sold to restaurants, and 47 percent of countryside CSAs sold to restaurants (USDA AMS 2017 CSA report).
The channel mix makes CSA farms look adaptable rather than rigid. Farmers markets and restaurants are common, but the mix changes by geography. That is useful for understanding why CSA can persist even when local competition, distribution, and consumer access patterns shift.
What demand and share data say
The USDA AMS survey asked operators about local-food demand, and the answers lean heavily positive.
- In the USDA AMS survey, 0.9 percent said local-food demand was declining significantly, 2.6 percent said it was declining somewhat, 11.3 percent said it was staying about the same, 60.3 percent said it was increasing somewhat, and 24.9 percent said it was increasing significantly (USDA AMS 2017 CSA report).
- 4 respondents rated local-food demand as declining significantly, 11 rated it as declining somewhat, 48 rated it as staying about the same, 257 rated it as increasing somewhat, and 106 rated it as increasing significantly (USDA AMS 2017 CSA report).
Demand was growing, but not evenly everywhere
The demand data matter because they show how operators perceived their market in 2014. The dominant view was growth: 60.3 percent said demand was increasing somewhat and 24.9 percent said it was increasing significantly (USDA AMS 2017 CSA report).
That growth perception lines up with the high share of sales coming from CSA itself. In the same survey, the average share of farm sales coming from the CSA was 53.2 percent, and 58.1 percent of surveyed CSAs said the CSA was the major market channel (USDA AMS 2017 CSA report).
How much of the farm did CSA represent?
The same survey also showed a spread in dependence on CSA revenue.
- 120 respondents said less than 25 percent of farm sales came from CSA (USDA AMS 2017 CSA report).
- 77 respondents said 25 to 49 percent of farm sales came from CSA (USDA AMS 2017 CSA report).
- 117 respondents said 50 to 74 percent of farm sales came from CSA (USDA AMS 2017 CSA report).
- 156 respondents said 75 to 100 percent of farm sales came from CSA (USDA AMS 2017 CSA report).
Those counts suggest that CSA is not only a supplementary channel. For a large share of operators, it is a primary revenue engine. The survey also found that 100 percent of CSAs used the CSA channel itself across all location groups (USDA AMS 2017 CSA report), which means the model’s core mechanism stayed universal even while the supporting channels varied.
What the survey signals about growth and difficulty
The survey does not describe CSA as effortless. It shows a market that is growing, but still demanding operational discipline.
- In the USDA AMS survey, 27.8 percent of already multi-farm CSAs supplemented new specialty products, 21.0 percent supplemented inventory when short, 15.9 percent cooperated on recruitment fairs, 14.4 percent cooperated on low-income voucher programs, and 11.4 percent cooperated on shared educational programs (USDA AMS 2017 CSA report).
- 11.2 percent said CSA shareholder recruitment was much less difficult in 2014 than previous years, 16.1 percent said it was somewhat less difficult, 46.9 percent said it was about the same, 20.3 percent said it was somewhat more difficult, and 5.5 percent said it was much more difficult (USDA AMS 2017 CSA report).
The operational story behind the numbers
Taken together, these figures show a category that is collaborative but still competitive. Some farms respond to complexity by sharing products, sharing recruitment, and sharing educational programs. Others lean on multiple sales channels to keep volume stable.
The recruitment numbers are especially telling. If 46.9 percent said recruitment difficulty was about the same, and another 25.8 percent said it was somewhat or much more difficult, then growth was not automatic even in a period where demand generally looked strong (USDA AMS 2017 CSA report).
That balance is the practical takeaway from the whole dataset. Community supported agriculture has a durable presence in U.S. direct marketing, but it works best when farms are operationally flexible, channel-diverse, and able to turn membership into a repeatable relationship.