Bonds Are Back

A broad de-risking out of equities usually hits the most crowded factor first, not the cheapest one. That means the near-term losers are likely the same AI-heavy large-cap growth names that have carried indices, because their ownership is already saturated and even a small allocation shift can compress multiples faster than earnings change. The cleaner relative winners are duration-sensitive fixed income proxies such as IEF/TLT and high-quality credit like LQD, while lower-quality equity beta and unprofitable growth are the first places where financing conditions tighten.

The second-order effect is important: a move from stocks into bonds is often less about recession and more about allocation fatigue after a concentrated run. In that case, the fundamental damage to AI infrastructure spend may be modest over the next 1-3 months because hyperscaler capex is planned annually and tends to lag sentiment. What can break the move is any renewed decline in real yields or a soft macro print that reinforces the "lower-for-longer" rate narrative; that would support both bonds and high-multiple tech, muting the rotation trade.

The contrarian read is that the market may be over-interpreting a portfolio rebalance as a regime shift. If earnings revisions for mega-cap tech stay positive, the AI complex can keep grinding higher even as some marginal capital migrates into bonds. The real tell is breadth: if leadership narrows further while credit spreads stay contained, this is mostly valuation rotation, not the start of a broader risk-off phase.