Market Research Report

A USDA-aligned evidence pack for perishable, time-sensitive produce — market scale, farm-size structure, logistics, shrink and food loss, and the definitions behind local sourcing claims — assembled for proposal-level positioning.

Scope and definitions

This Evidence Pack is tailored to a USDA Small Business Innovation Research Phase I proposal (Topic 8.6) and is restricted to perishable, time-sensitive produce (fresh vegetables, berries, fruit). It intentionally excludes storable commodity grains/oilseeds as the market dynamics (yields, logistics, shrink, and pricing) differ materially.

Primary national sources used (preferred by your requirements): U.S. Department of Agriculture agencies including USDA National Agricultural Statistics Service Census of Agriculture typology tables, USDA Agricultural Marketing Service transportation analysis (AgRTQ), and USDA Economic Research Service food loss documentation.

Farm size (“small” vs “large”) definition used in this pack (explicit and reproducible). This follows the 2022 Census of Agriculture Farm Typology table structure:

• Small family farms: Gross Cash Farm Income (GCFI) < $150,000 and $150,000–$349,999
• Midsize family farms: $350,000–$999,999
• Large family farms: $1,000,000–$4,999,999 and $5,000,000+
• Non-family farms: separate category in the Census typology table

“Local” definition (policy maximum vs practical service radius).

• USDA program “maximum” local definition (commonly used in federal procurement contexts): a product marketed <400 miles from origin or within the same State.
• Consumer perception benchmark (useful for website language): In a national consumer survey (late 2008), over 70% of respondents considered a 50‑mile radius “local.”
• For logistics calculations below, “local” is treated as 50–100 miles as a scenario, and “conventional” as 1,000–1,500 miles as a scenario (not asserted as a national average of “food miles”).

Key limitation you should disclose in your proposal narrative (important).

• The Census “vegetables” sales category used nationally is “Vegetables, melons, potatoes, and sweet potatoes.” Potatoes/sweet potatoes are included by Census taxonomy and are not always “time-sensitive” in the same way as leafy greens/berries. This pack uses that category as a best-available national proxy for perishable produce at scale, and flags the limitation everywhere it matters.

Replace/Correct These Website Claims

You didn’t provide your current website copy. To still meet your deliverable intent (“correct/ground claims”), below are common claims in local-perishable-produce ventures, with supported replacements (you can map them 1:1 to your actual text during final edits).

• Claim: “Small/local farms produce most of our vegetables (or fresh produce).”

Status: Unsupported (nationally false for produce sales). In Census typology, the two “small family farm” categories (<$350k GCFI) account for only ~4.1% of U.S. vegetable-group sales dollars (vegetables+melons+potatoes+sweet potatoes). Corrected claim (defensible): “U.S. produce sales are highly concentrated: very large family farms and nonfamily farms account for the majority of national produce sales.”

• Claim: “Small farms are more productive per acre than large farms (for fresh produce).”

Status: Not supported by this national Census proxy; direction is generally opposite in gross sales/acre for vegetables and berries. Using Census typology, gross sales per acre for the vegetable group are higher for very large family and nonfamily farms than for small-family categories. Corrected claim (defensible and precise): “Using USDA Census farm-typology data, produce sales-per-acre are higher on very large and nonfamily operations for vegetables and berries; small farms are more numerous but sell a smaller share of total volume/value.”

• Claim: “Local food means within 50 miles.”

Status: Partially supported (consumer perception), but not a universal USDA definition. Corrected claim: “USDA programs often use a maximum local definition of <400 miles or within-State, while many consumers perceive ‘local’ closer to ~50 miles.”

• Claim: “Reefer trucking costs are about $X per mile.”

Status: Supportable only if you cite a source and specify the freight context (spot produce lanes vs carrier operating cost). Corrected claim: “USDA AMS AgRTQ reported Q2 2024 produce truck rates of $6.45/mile (0–500 miles) and $2.81/mile (501–1,500 miles) (selected routes), while ATRI reports an all-truck average operating cost of $2.26/mile in 2024.”

• Claim: “Local sourcing cuts shipping cost by ~90%.”

Status: Conditionally supportable (depends on distance, rate band, and payload). Corrected claim (parameterized): “Using USDA AMS produce truck rates and a full-truck payload assumption, a 50–100 mile move can be ~0.8–1.6¢/lb, while 1,000–1,500 miles can be ~7.0–10.5¢/lb.”

• Claim: “Food waste is 40% of food.”

Status: Partially supported only with careful framing. USDA ERS’s loss-adjusted series estimates 31% loss at retail+consumer levels (2010) and notes exclusion of farm-level losses. Corrected claim: “USDA ERS estimated that 31% of the retail-level food supply went uneaten at the retail and consumer levels in 2010, excluding farm- and pre-retail losses.”

• Claim: “Fresh produce has the highest waste rates.”

Status: Supported for consumer-level loss shares and high retail shrink; quantify precisely. In ERS’s 2010 table, fresh fruit total loss = 37% and fresh vegetables total loss = 34%. Corrected claim: “In USDA ERS’s 2010 estimates, total losses were 37% for fresh fruit and 34% for fresh vegetables, with most losses at the consumer level.”

• Claim: “America’s food is cheap only because we ignore hidden costs.”

Status: Supportable if you cite a credible true-cost framework and state what’s included/excluded. Corrected claim: “A Rockefeller Foundation true-cost accounting analysis estimated the U.S. food system’s ‘true cost’ at ≥$3.2T/year versus ~$1.1T/year in food expenditures, with ~$2.1T in additional quantified external costs.”

Evidence tables

Table of farm scale vs outcomes for produce proxies

Interpretation guidance (important for reviewers): This table shows gross sales per acre (a value-per-acre proxy), not physical yields. It is derived from national 2022 Census typology tables and is among the best-available national size-segmented measures for produce-like categories.

Table 1: Farm scale vs outcomes (vegetable-group and berries) Sources: USDA NASS 2022 Census “Farm Typology” (sales + farm counts) and typology continuation pages (land in vegetables / land in berries). Calculated columns are shown again with step-by-step arithmetic in the next section.

Vegetables proxy = “Vegetables, melons, potatoes, and sweet potatoes” sales; denominator = “Land in vegetables” acres.

Berries proxy = “Berries” sales; denominator = “Land in berries” acres.

Reefer transport cost evidence and conversions

Important framing for reviewers: USDA Agricultural Marketing Service AgRTQ “truck rates” are observed produce lane pricing (market rates), while American Transportation Research Institute reports fleet operating cost per mile (carrier costs). These are related but not interchangeable (rates can be above or below costs depending on market conditions).

Table 2: Reefer transport costs (rate → $/lb-mile → $/lb) Sources: AMS AgRTQ Q2 2024 national summary truck-rate table; ATRI operating-cost metric via TRID record abstract. Assumption for conversion: 40,000 lb net payload (user-adjustable; see sensitivity table and formulas). Rates are per mile; conversion is purely arithmetic.

Representative distance conversion (example “local” vs “conventional”) using AMS AgRTQ Q2 2024 rates and 40,000 lb payload.

Sensitivity to payload (illustrative; plug in your actual average net payload). Using the same AMS rates above.

Produce waste and shrink evidence

Table 3: Waste/shrink (supply chain + on-farm field loss evidence) Core USDA food-loss benchmark (2010): ERS Loss-Adjusted Food Availability (LAFA) table for retail and consumer losses by commodity group. Retail shrink detail: ERS Food Availability System “Food Loss” page (2016 retail shrink volumes and ranges for fresh produce). On-farm: field measurement study of NC vegetable farms (unharvested/unused produce volumes).

USDA’s national value-of-loss benchmark (often cited in proposals): ERS estimated total food loss at retail + consumer levels in 2010 at ~133B lb (31%) and $161.6B (retail value), explicitly excluding farm-level and farm-to-retail losses.

Health and environmental externality estimates

Table 4: “True cost” externality estimates (U.S.-specific, with scope statements) Primary source: The Rockefeller Foundation True Cost of Food framework (2021).

Scope caveat (use verbatim in your proposal if needed): The report’s U.S. food system cost analysis includes production through consumption stages but states it does not include food service/hospitality in cost analyses.

Reproducible calculations and dataset pulls

Reproducible calculations

All arithmetic below is computed directly from the cited sources.

Calculation set for productivity proxy: sales per acre by farm typology

1. Vegetable-group sales per acre for “very large family farms (GCFI ≥ $5M)”

Inputs from Census typology:

• Vegetable-group sales = 14,705,441 ($1,000) = $14,705,441,000
• Vegetable acres (“land in vegetables”) = 1,771,631 acres

Formula: Sales per acre = (Sales dollars) / (Acres)

Arithmetic: $14,705,441,000 ÷ 1,771,631 = $8,300.51 per acre

1. Vegetable-group sales per acre for “small family farms (GCFI < $150k)”

Inputs:

• Sales = 728,484 ($1,000) = $728,484,000
• Acres = 131,417 acres

Arithmetic: $728,484,000 ÷ 131,417 = $5,543.30 per acre

1. Berries sales per acre for “nonfamily farms”

Inputs:

• Berry sales = 2,100,211 ($1,000) = $2,100,211,000
• Berry acres (“land in berries”) = 94,984 acres

Arithmetic: $2,100,211,000 ÷ 94,984 = $22,111.21 per acre

Calculation set for shipping overhead: $/lb from $/mile

1. Cost per lb for a 1,000-mile shipment (conventional scenario)

Inputs:

• Rate = $2.81/mile (AMS AgRTQ, Q2 2024, 501–1,500 miles band)
• Distance = 1,000 miles (scenario assumption)
• Payload = 40,000 lb (scenario assumption; adjust)

Formula: Cost per lb = (Rate $/mile × Distance miles) ÷ Payload lb

Arithmetic: ($2.81 × 1,000) ÷ 40,000 = $2,810 ÷ 40,000 = $0.07025/lb = 7.025¢/lb

1. Cost per lb for a 50-mile shipment (local scenario)

Inputs:

• Rate = $6.45/mile (AMS AgRTQ, Q2 2024, 0–500 miles band)
• Distance = 50 miles (scenario assumption)
• Payload = 40,000 lb (scenario assumption)

Arithmetic: ($6.45 × 50) ÷ 40,000 = $322.50 ÷ 40,000 = $0.00806/lb = 0.806¢/lb

1. Implied $/lb-mile (useful for sensitivity and for proposal clarity)

Formula: $/lb-mile = ($/mile) ÷ (payload lb)

Example (0–500 band): $6.45 ÷ 40,000 = $0.00016125 per lb-mile

Calculation set for market sizing: population × per-capita produce spending

1. U.S. per-capita produce consumer sales (range)

Inputs:

• U.S. consumer sales of fruit+vegetable products estimated at $160B–$190B annually (CRS estimate). Congressional Research Service
• U.S. population (July 1, 2024) = 340.1 million from U.S. Census Bureau.

Per-capita (low) = $160,000,000,000 ÷ 340,100,000 = $470.45/person/year Per-capita (high) = $190,000,000,000 ÷ 340,100,000 = $558.66/person/year

1. Ohio produce-market proxy (consumer sales)

Inputs:

• Ohio population (July 1, 2024) = 11,883,304.
• Per-capita range from step 7.

Low: 11,883,304 × $470.45 = $5.59B/year High: 11,883,304 × $558.66 = $6.64B/year

1. Summit County proxy (consumer sales)

Inputs:

• Summit County population (July 1, 2024) = 538,370.

Low: 538,370 × $470.45 = $253.3M/year High: 538,370 × $558.66 = $300.8M/year

Bash download commands for key public artifacts

The commands below create a local folder and download each key PDF used in the tables above.

mkdir -p evidence_pack/sources
cd evidence_pack/sources

# USDA NASS 2022 Census of Agriculture - Farm Typology (United States)
curl -L -o usda_nass_2022_farm_typology_us.pdf \
 "https://www.nass.usda.gov/Publications/AgCensus/2022/Online_Resources/Typology/typology_us.pdf"

# USDA AMS Agricultural Refrigerated Truck Quarterly (Quarter 2, 2024)
curl -L -o usda_ams_agrtq_q2_2024.pdf \
 "https://www.ams.usda.gov/sites/default/files/media/RTQ2ndQuarter2024.pdf"

# USDA ERS LAFA documentation (HTML; save as a file for audit trail)
curl -L -o usda_ers_lafa_documentation.html \
 "https://www.ers.usda.gov/data-products/food-availability-per-capita-data-system/loss-adjusted-food-availability-documentation"

# USDA ERS Food Loss page (HTML; includes fresh produce retail shrink findings)
curl -L -o usda_ers_food_loss_page.html \
 "https://www.ers.usda.gov/data-products/food-availability-per-capita-data-system/food-loss"

# CRS fruit & vegetable market sizing / pricing report (includes the $160B–$190B estimate method)
curl -L -o crs_marketing_pricing_us_fruit_veg_industry.pdf \
 "https://www.congress.gov/crs_external_products/R/PDF/R48213/R48213.2.pdf"

# Rockefeller Foundation true cost report
curl -L -o rockefeller_true_cost_food_full_report.pdf \
 "https://www.rockefellerfoundation.org/wp-content/uploads/2021/07/True-Cost-of-Food-Full-Report-Final.pdf"

Ready-to-cite sentences

Each sentence is ≤25 words and designed to drop into your SBIR narrative or website.

• In 2022, very large family farms (GCFI ≥$5M) generated 52.2% of U.S. vegetable-group sales.
• Small family farms (GCFI <$350k) generated 4.1% of U.S. vegetable-group sales in 2022.
• In 2022, U.S. vegetable-group sales totaled $28.2B (vegetables, melons, potatoes, sweet potatoes).
• USDA ERS estimates fresh fruit losses totaled 37% in 2010 (12% retail, 25% consumer).
• USDA ERS estimates fresh vegetable losses totaled 34% in 2010 (10% retail, 24% consumer).
• AMS AgRTQ reported Q2 2024 produce truck rates of $6.45/mile (0–500 mi) and $2.81/mile (501–1,500 mi).
• Using AMS Q2 2024 rates and a 40,000-lb payload, 1,000 miles implies ~7.0¢/lb in linehaul trucking cost.
• ERS estimates 133B lb (31%) of the 2010 food supply went uneaten at retail and consumer levels, excluding farm-level losses.
• A consumer survey found 82% of respondents had bought locally grown fresh produce; over 70% considered food grown within 50 miles “local.”
• A true-cost accounting analysis estimated the U.S. food system’s “true cost” at ≥$3.2T/year versus ~$1.1T/year in expenditures.

Limitations and how to position this evidence in a USDA SBIR Phase I market analysis

Why these are “best available” and not perfect. National data rarely report physical yields by farm size for “fresh vegetables/berries/fruit” as a single category. The 2022 Census farm typology is one of the few national sources that jointly provides:

• consistent, national farm-size segmentation (by GCFI),
• commodity-group sales values,
• and crop land-use measures (e.g., land in vegetables; land in berries).

How to present the productivity story credibly (and reviewer-proof).

• Say “sales per acre” (gross value/acre) rather than “yield,” unless you later add a commodity-specific yield dataset.
• Explicitly call the vegetable category a proxy and note it includes potatoes/sweet potatoes.
• Emphasize that your innovation targets time-sensitivity and shrink, where value is lost even when production is high; align this with the ERS finding that consumer-level loss is large for fresh produce.

How to position logistics cost without overclaiming.

• It is defensible to say that trucking adds single-digit cents per pound locally and high single- to low double-digit cents per pound for cross-country distances, given explicit assumptions.
• Also note that these are linehaul rate conversions and exclude additional real-world costs (multi-stop routing, detention, rejected loads, temperature excursions, empty backhaul, etc.), which you can list qualitatively without inventing numbers.

How to position the “waste problem” without mixing incompatible scopes.

• Use ERS LAFA for a headline national benchmark (31% across all foods; 34–37% for fresh produce groups).
• Use field-measured on-farm evidence (e.g., 42% of marketed yields in NC vegetable fields) as contextual evidence that farm-level loss can be substantial and often missing from national accounting.
• Clearly state that the ERS 2010 national totals exclude farm-level and pre-retail losses.

How to use “true cost” in SBIR without triggering reviewer skepticism.

• Lead with the scope statement (“does not include food service/hospitality in cost analyses”) and that it is a lower bound (“at least $3.2T”).
• Present the quantified domains (health/environment/biodiversity/livelihoods) as externalities context and keep your core grant logic anchored in measurable Phase I outcomes (reduced shrink, reduced miles, reduced handling time, improved cold-chain reliability).