Market Correction Risk Rising As Midterm Madness Approaches | Lance Roberts

Summary of the video’s main points

• Risk of a market correction approaching midterms (5–10%)

  • Lance Roberts argues that if semiconductors keep sliding and investors face margin calls, that pressure can spread to the broader market.
  • He highlights August/September as the highest-risk window and suggests a correction of ~5–10% into the midterms as the “most likely” outcome over roughly the next two months.
  • In that kind of environment, he expects riskier areas to underperform first, followed by potentially better buying opportunities later—into November/December, which he describes as seasonally stronger.

• Tech volatility is not necessarily a systemic “panic” (possible breadth/rotation)

  • Adam asks whether tech cooling reflects real concern; Roberts replies that the signals point to rotation and broadening, not just a narrow selloff in tech.
  • While MAG 7 and semiconductors are down on headlines (including AI competition chatter), Roberts claims other segments look comparatively better:
    • Energy (especially after renewed Iran “re-ignition” concerns)
    • Banks / investment banking results
    • Broader measures like an equal-weight index
  • He attributes some of the day-to-day movement to options expiration and momentum unwinds.

• Margin/positioning concerns tied to semiconductors

  • Roberts emphasizes margin debt and leveraged retail exposure concentrated in semiconductors (including references to weakness in the KOSPI semiconductor index and similar areas).
  • If semiconductor leverage/liquidations spread, it could worsen market breadth and reinforce correction odds.

• Portfolio positioning: defense/offense shift and hedging

  • Roberts describes a model-driven approach:
    • Already buying S&P puts (he says he increased/added puts “yesterday morning” in a platinum/options model).
    • Raising cash / reducing equity exposure in upcoming steps if conditions worsen.
    • Using a money flow and breadth indicator in his weekly newsletter process to decide when to stay exposed vs. cut risk.
  • He says the portfolio is currently positioned for risk-off rotation:
    • shifting from offense (hot semis/tech momentum) toward defensive/quality factors
    • with the defensives outperforming recently

• Technical backdrop: consolidation, support test, and potential breakdown

  • Roberts presents a chart-based view:
    • The market has been consolidating since June, with “rising bottoms.”
    • It tested the 50-day moving average and bounced, but he warns it could still break down.
  • He links correction risk to weakening momentum, reiterating that semiconductor weakness + margin calls is the likely transmission channel.

• Earnings and forward estimates as the ultimate driver

  • A central thesis: markets ultimately reprice based on earnings growth and forward earnings estimates.
  • He notes that semiconductor “sensor”/analyst forecasts seem to show a slight negative revision trend (“wiggle down”) recently, aligning with weaker semiconductor performance.
  • While he treats one-week forecast changes as noise, he argues that a sustained pattern of negative revisions would be more meaningful.

• Debate: AI valuations and claims that “earnings are inflated”

  • Adam discusses Fred Hickey’s bearish AI stance:
    • AI hyperscaler earnings may include one-time gains tied to frontier model milestones
    • a “depreciation schedule wave” could reduce forward profitability
    • adjusted for this, implied P/E could be far higher than investors assume (Fred cites averages around ~67 P/E after adjustments)
  • Roberts’ response:
    • Markets are dynamic and will “look through” accounting effects as reality is reflected in earnings revisions
    • The key question is whether revenue growth keeps beating cost growth / capex needs
    • He pushes back on the idea that mega-cap firms are “blindly throwing money” into AI without monetization planning

• AI cost/productivity concerns acknowledged but treated as early/uncertain

  • Adam cites an example: a major AI investor/podcast guest claims token spending (compute costs) doubled every ~45 days while productivity gains were modest (~5%), implying costs could outpace near-term incremental output.
  • Roberts counters:
    • it’s one anecdote
    • avoid firm conclusions without broader evidence
    • but he agrees the industry should be monitored for signs that revenue growth slows

• Rotation “baton” across factors (growth/value/momentum)

  • Roberts explains factor-model classifications carefully:
    • “Growth” vs “value” in their framework is based on PEG / earnings growth / valuation, not just price movement.
  • He says the model can shift between value and growth depending on relative performance.
  • His framework may flip allocations if leadership changes (e.g., semis become oversold → rotate back; if value becomes crowded → rotate elsewhere).

• Commodities segment: gold miners and energy

  • Gold miners (GDX)
    • Roberts says gold miners are extremely stretched after a parabolic move and likely face more downside before long-term mean reversion.
    • Even when miners are profitable at high gold prices, miners can become “dead money” if sentiment doesn’t rotate back.
    • He suggests picking individual names may be safer than broad miner exposure, because miners remain commodity dependent.
  • Energy
    • He favors energy, especially natural gas and pipelines; he notes oil has moved from oversold to more overbought.
    • He says the market’s reaction to Iran has been muted so far relative to worst fears.
    • Oil catalysts matter: headlines, weather (summer heat / El Niño), and escalation vs. de-escalation.
    • He argues broader earnings may have already priced much of the Iran impact—unlike earlier “exogenous shock” phases.

• Bonds/macro: rates steady; inflation not the “doom scenario”

  • He notes some weak spots in sentiment (e.g., BDC-related credit), but claims broader high-quality credit shows little built-in risk premium.
  • 10-year yields are roughly flat around ~4.5%, trading sideways rather than breaking higher.
  • He argues inflation is likely muted unless oil spikes dramatically and persists.
  • Shelter/housing disinflation is a key point; he acknowledges housing pressure in some markets but does not frame it as a crash.

• Practical risk management advice

  • Roberts emphasizes:
    • don’t overreact to headlines or every correction
    • avoid “stupid risks” like margin, leveraged ETFs, or aggressive short-dated options
    • invest with a plan and focus on long-run earnings fundamentals
  • He also recommends working with advisers who have lived through real bear markets (not just mild corrections), because psychology becomes the actual risk.

Presenters / contributors

• Adam Tagert (host; “Thoughtful Money,” founder and host)
• Lance Roberts (electorally suspicious portfolio manager; guest)
• Fred Hickey (referenced as an interview subject whose bearish AI views are discussed)
• Tom Hoenig / Lacy Hunt / Judy Shelton (mentioned as planned future panelists; not contributors within the current discussion)