Nurseries face a sustained U.S. labor shortfall that H‑2A visas only partly fill; automation has inched up about 15 percentage points over 15 years but remains patchy—no task is majority-automated—so public funding and shared-equipment models will be pivotal to develop weather-robust, low-cost AI/robotics for this heterogeneous sector.

Main Finding

The U.S. nursery (green) industry is facing a sustained labor deficit that current labor markets and immigration (H-2A) responses only partially mitigate. Automation adoption has increased modestly (~15% overall adoption growth over 15 years to 2021), but uptake remains fragmented and no single nursery task is automated by a majority of firms. Given the small, heterogeneous, and outdoor nature of nursery operations, private R&D alone is unlikely to fully solve the labor shortfall — public funding (USDA, NSF) and new business models will be important to spur AI/robotics solutions that are robust to weather and site heterogeneity.

Key Points

• Labor shortage:
  • Prime-age employment in the U.S. is high (~80% for ages 25–54), yet nurseries report inability to hire due to (a) wage pressure and (b) unqualified local applicants.
  • Falling U.S. birth rates mean a smaller domestic labor pool for the coming decades, so demographic relief is unlikely.
• Immigration stop-gap:
  • Usage of the H-2A temporary agricultural visa for the green industry has risen sharply over the past decade, helping fill gaps.
  • H-2A has limitations: providing housing, training, and meeting wage expectations is costly; H-2A workers can be more expensive than local seasonal labor and come with administrative constraints.
• Automation trends:
  • Percent of all tasks automated increased by roughly 15% over the 15 years ending 2021.
  • No single task was automated by more than 57% of respondents in 2021 (up from 52% in the mid-2000s), indicating dispersed and partial automation rather than task-wide substitution.
• Market characteristics:
  • Nursery operations are a niche, often small-scale industry with non-uniform operating environments and exposure to weather extremes — both complicate robotics/automation design.
  • These characteristics make nursery crops a specialized opportunity for R&D, but also a risky market for private investors without public support.
• Policy & funding environment:
  • Federal agencies (USDA, NSF) have increased funding for automation in specialty crops, creating pathways for innovators to develop suitable technologies.

Data & Methods

• Primary evidence summarized from:
  • A 2021 national labor survey of nurseries which reported adoption rates and task-specific automation shares (57% max for any one task in 2021; 52% in mid-2000s).
  • Comparative adoption trend: ~15% increase in percent of all tasks automated across a 15-year span to 2021.
  • Industry reports and administrative data indicating large increases in H-2A usage for the green industry over the past decade.
  • Public funding announcements and program activity from USDA and NSF signaling greater emphasis on specialty-crop automation.
• Methodological notes and limitations:
  • Evidence is primarily observational and survey-based; causal inferences about automation reducing labor shortages are limited.
  • Survey aggregates mask heterogeneity by firm size, region, crop type, and capital access.
  • Specifics on sample sizes, survey design, and firm-level economics (costs of automation vs. labor) are not provided here — needed for detailed cost–benefit or adoption-modeling work.

Implications for AI Economics

• Market size and R&D incentives:
  • Nursery crops are a small, heterogeneous market: private R&D may underinvest due to limited scale and high engineering complexity. Public funding (grants, challenge prizes) can correct this market failure by reducing early-stage risk.
• Technology design constraints:
  • Research should prioritize robustness (weather tolerance), modularity (fits diverse nursery layouts), and low capital cost or shared-asset models (leasing, co-ops) to improve adoption among small firms.
• Adoption economics and labor impacts:
  • Partial automation (task-level) suggests many technologies will be complementary to workers rather than full substitutes — potential to raise productivity and wages for retained workers, but also to change skills demanded.
  • Policymakers should anticipate reallocation/frictions: training, certification, and extension services can speed diffusion and improve labor quality.
• Immigration and dynamic substitution:
  • H-2A expansion is a stop-gap; long-term reliance on seasonal visas affects incentives for automation (if cheap labor is readily available, investment may be delayed). Policy design (e.g., visa costs, housing requirements) will influence the automation-business case.
• Public policy levers to accelerate useful automation:
  • Targeted R&D grants for outdoor-robust robotics and sensing.
  • Matching funds or tax credits for adoption by small nurseries.
  • Support for shared-equipment models (regional robotics hubs, leasing) and demonstration projects to reduce adoption risk.
  • Workforce-development programs pairing automation deployment with retraining and upskilling.
• Research opportunities for AI economists:
  • Cost–benefit analyses comparing automation vs. H-2A and local labor across regions and crop types.
  • Models of adoption under seasonal demand and heterogeneous firm sizes.
  • Evaluation of public R&D interventions (funding, demonstration projects) on adoption, productivity, and employment outcomes.
  • Study of complementarities between AI/robotics and human labor in non-uniform, weather-exposed agricultural contexts.

Summary takeaway: Nurseries face a persistent labor supply problem that has prompted greater use of H-2A and modest automation gains. Because nursery operations are small-scale and heterogeneous, public funding and creative business models will be crucial to mobilize AI/robotics solutions that are robust, affordable, and complementary to existing workers.