In the age of AI and late-career layoffs, a second act can be a financial hedge

RBC Economics data shows workers aged 45+ accounted for nearly 40% of the increase in unemployment between June 2024 and June 2025, highlighting late-career vulnerability. The article profiles a 59-year-old who retired five years early, accepted a smaller pension, but rebuilt income through psychotherapy and Reiki to roughly match her college salary while keeping the pension. Policymakers and planners should note available tax support for re-skilling: the Lifelong Learning Plan allows withdrawals up to $10,000/year (max $20,000) tax-free if repaid within 10 years. Advice: build second-act skills and side income while employed, and factor upskilling costs into retirement plans to mitigate AI- or tariff-driven job disruption.

Analysis

An aging, mid-career workforce intentionally building “career optionality” is not just a social trend — it changes labor supply composition and pricing power across knowledge-intense services. Expect a steady increase in supply of high-experience, part-time consultants and therapists that reduces marginal rates for junior consulting gigs while expanding employer access to senior talent at sub-FTE economics within 12–36 months.

Training and marketplace platforms that connect experienced talent to buyers (career platforms, gig marketplaces, and employer-integrated reskilling vendors) gain durable network effects: more experienced sellers attract more buyers, increasing platform take-rates and average ticket sizes. Conversely, generic course aggregators or credential mills without employer placement funnels risk commoditization as credential inflation rises and buyers prioritize demonstrable on-the-job outcomes.

Tax and policy levers (e.g., education withdrawal programs or expanded training tax credits) are key amplifiers — a modest expansion of subsidy or repayment flexibility could accelerate demand for paid upskilling by 20–40% over 2 years. The wildcard is AI: it both drives need for reskilling and competes with mid-skill advisory roles; if large language models replace routine advisory tasks within 18–36 months, the composition of second acts will tilt toward uniquely human services (mental health, governance, creative consulting).

Monitor fee compression in independent advisory markets and rising supply of retirees choosing flexible second careers; these are leading indicators that platform pricing power is being tested even as volumes climb.

Key Decisions for Investors

• Long UPWK (Upwork) — 12–24 month horizon. Rationale: platform captures growing pool of senior part-time consultants and benefits from higher take-rates as enterprises source fractional expertise. Positioning: buy shares or buy 18–24 month call spreads to limit downside; reward skew positive if marketplace GMV growth accelerates >20% YoY. Key risk: continued fee compression and competition from direct-hire models.
• Long COUR (Coursera) and TWOU (2U) pair — 12–36 months. Rationale: favor providers with employer integration and degree/career outcomes over pure content libraries. Trade: long COUR, sell a small hedge in CHGG (Chegg) or a generic course aggregator to express outcome-focused education vs commoditized tutoring. Risk/reward: high if enterprise sales and cohort model adoption expand; downside if credential inflation reduces willingness to pay.
• Long TDOC (Teladoc) or mental-health specialist exposure — 18 months. Rationale: second-act trend increases demand for counseling/wellness services and telehealth monetizes flexible schedules. Implementation: buy equity or layered LEAPS calls; watch reimbursement/regulatory changes as primary catalysts.
• Cautious short / avoid: legacy for-profit colleges and non-integrated vocational schools — 12–24 months. Rationale: they lack employer placement and will face credential skepticism and subsidized, outcome-driven competitors. Consider pair trades: short selected education names versus long outcome-linked providers.
• Event trigger sizing rule: increase exposure (add 25–50%) if: (a) government expands training tax credits or LL program within 6–12 months, or (b) quarterly metrics show >15% growth in mid-career segments (45+) on platform user cohorts. Exit/hedge if platforms report accelerating take-rate compression or AI-driven automation materially reduces hours billed per consultant (>10% QoQ).