How recent grads are dealing with the shrinking pool of entry-level jobs: 'None of us are really thriving'

The labor market for recent college graduates is weakening: a Cengage survey found only 30% of the more than 2 million 2025 bachelor’s graduates reported a full-time job in their field, and 76% of employers hired the same number or fewer entry-level workers in 2025 versus 2024. Downward hiring, 1.1 million announced job cuts Jan–Oct 2025 (Challenger) and a 9.7% unemployment rate for recent grads as of Sept. 2025 (Federal Reserve) reflect firms pulling back amid inflationary pressures, tariff changes and the adoption of AI that is slowing new hires. The trend is pushing some young workers toward trades—where wages and BLS growth projections are stronger—potentially reshaping early-career earnings trajectories and long-term wage growth.

Analysis

Market structure: Slower entry-level hiring and AI-driven headcount pacing favor capital-light, productivity-enhancing tech and trade-oriented goods over broad-service discretionary demand. Near-term winners: home-improvement and tools (DIY demand, HD resilient), vocational/training providers and industrial equipment suppliers; losers: hospitality/entry-level-heavy retail and gig-heavy consumer services (MAR exposed), with a 6–12 month drag on wage-driven consumption that can shave ~0.1–0.3pp off GDP growth if persistent.

Risk assessment: Tail risks include rapid, broad AI adoption that structurally reduces entry-level roles (10–25% cohort displacement scenario), and regulatory pushback on AI hiring practices. Immediate shocks (days–weeks) are sentiment-driven; medium term (3–9 months) depends on payroll reports and Fed reaction; long term (1–3 years) is human capital scarring reducing lifetime earnings and consumption. Hidden dependency: lower young-worker mobility suppresses wage inflation, which could prompt earlier Fed easing if unemployment widens by +50–75bps.

Trade implications: Implement defensive longs in HD and select industrials, paired with short exposure to high fixed-cost hospitality (MAR); use options to cost-effectively express downside in lodging. Fixed income: tactical buy of long-duration Treasuries (TLT) if unemployment for 20–24s breaches 10.5% for two consecutive months or monthly NFP <+100k, signaling Fed pivot. Rotate portfolio overweight to durable goods/skills training and underweight consumer discretionary and cyclical services for next 6–12 months.

Contrarian angles: Consensus understates the multi-year re-rating of industrials and training providers as labor re-skilling accelerates—this is underbought. Reaction to college-value narrative may be overdone; education/upskilling platforms (public names) could see durable revenue tailwinds if underemployment persists. Also, weaker early-career wages can be disinflationary, creating a late-2026 reflation trade if wages snap back, so keep nimble hedges.

Key Decisions for Investors

• Establish a 2–3% long position in Home Depot (HD) over the next 2 weeks as a defensive DIY/durable-goods play; trim if quarterly comps fall below +2% y/y or if same-store sales miss consensus by >200bps.
• Initiate a 1–2% short position in Marriott (MAR) via a 3-month put spread (buy 15% OTM, sell 30% OTM) sized to 1% portfolio risk, reflecting near-term demand sensitivity and headcount exposure; reassess after next two monthly travel demand prints.
• Allocate 2% to long-duration Treasuries (TLT) if: US unemployment rate for 20–24-year-olds >10.5% for two consecutive months OR monthly nonfarm payrolls <+100k, as a trigger for potential Fed easing; hedge equity beta by 50% when deploying.
• Rotate +3% net into industrials/skills: add 1–2% positions in industrial-equipment or electrical contractors (e.g., CAT or relevant ETF) and 1% in education/upskilling public plays (e.g., CHGG or MAN), targeting 6–12 month horizon for re-rating as trades hiring tightens.
• Use a pair trade: long HD (2%) / short MAR (1%) to express durable-goods resilience vs. hospitality sensitivity; rebalance after 3 months or after material labor-market data divergence (>50bps change in unemployment).