← Analysis page  ·  Paulo Macro hub  ·  Research hub

Actionable insights — Fly on the Wall with Shrub & Cloudbear

Not what they own, but how they reason: size a hated contrarian macro bet as a defined-risk option and judge it on payoff asymmetry, not probability; read a mega-IPO calendar as a two-sided supply signal; and check whether fully-funded pensions and high real yields are quietly building a structural bid for bonds.
2026-SEP-10 · PauloMacro Substack · paid audio conversation with Le Shrub — Substack player positions, no deep-links · ↗ Listen / read original · full analysis · transcript
How to read this page: each insight is a method — the data you pull, the diagnostic question, and the signal to watch when re-running it. The boxed line shows how it played out in this conversation. Speaker attributions are inferred (the transcript carries no labels); m:ss = Substack audio position.

22:091. Put the hated contrarian trade on as capped-loss optionality and grade it on payoff

The repeatable method
  1. Identify the macro trade that is openly mocked — count the tells: TV personalities, a magazine cover, famous former bulls capitulating to the other side.
  2. Name the mechanism that could make the consensus wrong (here: an official sector that has already suppressed two other markets and has tools left for this one).
  3. Express it only with premium-limited options, and fix the total risk as a small portfolio percentage before entry (~0.3%) so a loss is known and tolerable.
  4. Write down rough odds for each outcome and the payoff multiple; take it only if probability × payoff clearly beats the premium even at low odds.
  5. Keep the structural view separate — the trade can run while the long-term thesis stays hostile (he owns no bonds otherwise).
Here:
Le Shrub's TLT option position: 30bps at risk, ~10× if it works; ~50-50 yields hold, 20-30% they fall; tells = Cramer, a Barron's cover, Lacy Hunt and Dalio bond-bearish; mechanism = Bessent "two out of two" on crude and the yen.
Watch for

14:382. Treat a mega-IPO in the calendar as a two-sided signal

The repeatable method
  1. Before a very large listing, expect the sell side to push supportive narratives and headline "breakthroughs" — discount the news flow accordingly.
  2. List what could derail the window: a cheaper competitor release, political or regulatory pressure, weak semis/sector tape, an election that shifts regulation risk.
  3. If the deal prices, ask whether it marks the top of the theme (the supply is the peak).
  4. If it is pulled or delayed, read that as a likely near-term market low: the supply overhang that investors had been raising cash for disappears.
  5. Weight the calendar: the closer to year-end, the less appetite managers have to risk their year on a controversial deal.
Here:
Anthropic's IPO at a ~$2tn talk price: Le Shrub says it gets done and may be "the absolute meme top"; Paulo cites DeepSeek's new model at 1.4% of the cost, congressional letters to OpenAI and 4-5:1 data-center opposition — a pulled deal = "a near-term low in the market."
Watch for

28:173. Check pension funding status against long yields for a slow structural bond bid

The repeatable method
  1. Track corporate pension funded ratios (asset gains plus the higher discount rate shrinking liabilities).
  2. Compare the plans' liability discount rate with what they can lock in on long Treasuries and TIPS.
  3. When plans are >100% funded and long yields match or exceed their discount rate, expect gradual de-risking — equities sold, long bonds bought.
  4. Use the emerging-market template (Brazil): high real yields anchor domestic money in sovereign debt, and equities only rally when real yields compress.
  5. Treat it as background, not a timing tool: it slows a bond sell-off and turns a dollar collapse into a slow bleed.
Here:
Paulo: US plans went from 60-70% funded post-GFC to "completely defeased and more than 100% covered"; with ~5.5% on the 30-year or TIPS at ~2.5% real, "you do it" — "one more way we are actually speaking Portuguese."
Watch for

Methods distilled from the PauloMacro Substack paid audio conversation of 2026-SEP-10 for personal study. Not investment advice.