← Analysis page  ·  James Davolos hub  ·  Research hub

Actionable insights — Rates Aren't Coming to Save You

The repeatable analysis behind the picks: not what he owns, but how he found it — written so the process can be rerun later on different names.
2026-SEP-12 · Market Talk with George Noble · James Davolos (Horizon Kinetics) · ▶ Watch · full analysis · transcript
How to read this page: each insight is a method — the trigger that put him onto an idea, the steps that turned it into a position, and the signal to watch when re-running it. The boxed line shows how it played out in this appearance. Timestamps deep-link into the video.

5:10 1. Regime check — find the hidden rate bet in a "bottom-up" portfolio

The repeatable method
  1. For each holding, estimate how much of today's value sits in the terminal value (for a non-cash-generating business, ~90%).
  2. Re-price it at the discount rate the current regime implies, not the 3–5% of 2010–19.
  3. If the thesis only works when rates return to zero ("I just need the macro to come back"), label it a macro/factor bet, not a stock pick — and size it accordingly.
Here: the 40×-revenue compounder crowd waiting for a rates "savior" vs his camp positioning for a new capital cycle and cost of capital (7:06).
Watch for

12:12 2. Fiscal arithmetic → pick the inflation regime by elimination

The repeatable method
  1. Add mandatory spending + interest expense and compare to federal revenue; when it nears 100%, every discretionary dollar is debt-funded.
  2. List the exits — hard default, austerity, productivity miracle, nominal growth/debasement — and test each: default is unthinkable, austerity triggers recession (and bigger deficits), productivity needs > the internet's ~3%/yr with a flat labor force.
  3. Whatever survives sets the portfolio regime; here, "let it run hot" → own nominally indexed cash flow.
Here: Cembalest's cliff + GDP = labor × productivity math → high nominal growth and debasement "really is the only way out" (14:28, 15:58).
Watch for

17:31 3. The nominally-indexed-cash-flow screen (real asset × capital light)

The repeatable method
  1. Require a scarce, tangible real asset with pricing power (the revenue side indexes to inflation).
  2. Require a capital-light model where costs don't grow with revenue (the expense side doesn't) — score it on TOLL: tangible asset, oligopoly, low incremental capital intensity, long-duration cash flow.
  3. Search where both coexist: royalties (energy, metals), raw land, water infrastructure, exchanges/brokerages.
Here: the Stahl real-asset capital-light framework (9:00) → royalties PBT, WHK, FNV; the fund itself INFL.
Watch for

19:22 4. Own the royalty, not the producer — judge the downside, not the upside

The repeatable method
  1. Don't pick by upside leverage (the highest-cost producer always wins that); pick by whether the business survives the down-cycle without issuing debt or equity.
  2. Check reinvestment behavior: producers reinvest peak cash flow at peak prices into higher break-evens (the capital cycle); royalties don't.
  3. Value the non-producing assets separately — sell-side NAVs usually count only producing assets, so those options are free.
Here: ~90% gross / 70–80% operating margins; FNV's unvalued non-producing tail kicks in if gold holds the $4,000s (41:49).
Watch for

25:23 5. Stress-test "this time is different" margins with the capital cycle

The repeatable method
  1. When a sector's margins are capitalized into perpetuity, ask who is incentivized to add supply.
  2. Give extra weight when the would-be new supply is the incumbent's own best-capitalized customers — they can self-supply to avoid the premium and bottlenecks.
  3. Assume mean reversion eventually; the only question is timing (it "could go on for longer than people expect").
Here: GOOGL and AMZN building their own chips to dodge the "Jensen tax" → caution on NVDA-style margins (25:45).
Watch for

31:53 6. Special-situation royalties — structure changes and consolidators

The repeatable method
  1. Look for structure upgrades: a net-profits interest (paid after opex/capex) converting to a straight royalty, especially when an activist/long-term holder drives it.
  2. Look for unvalued adjacent assets attached to the restructuring — surface land, water handling, power substations, data-center sites.
  3. Or look for fragmented royalty markets with one natural scale buyer: the consolidator can compound through accretive bolt-ons others can't do.
Here: PBT (SoftVest/Blackbeard NPI → NPRI + land) and WHK (the only scale buyer of fragmented Marcellus/Utica/Haynesville gas royalties, 36:54).
Watch for

37:56 7. Underwrite a base rate at normalized prices, then add "turns"

The repeatable method
  1. Model only visible organic cash flow at normalized commodity prices (below spot for elevated oil, around spot for gas) — no acquisitions, no macro help.
  2. Require that alone to clear an 8–12% base return.
  3. List the optional "turns" separately — inflation lifting normalized prices, land/water/power monetization, accretive M&A — and treat mid-teens to low-20s as the upside case, never the entry case.
Here: PBT and WHK both underwritten at a 10–12% minimum with mid-teens/low-20s possible (39:17).
Watch for

27:59 8. Split the energy call: oil, gas, and the basis differential

The repeatable method
  1. Analyze oil and gas separately — different end markets, transport and refining.
  2. For oil, ignore barrel-counters and model the swing buyer's behavior (China's import and refinery-run decisions).
  3. For gas, find basins where local gas trades at a deep discount or negative, then ask what could close the gap (new LNG export capacity, in-basin power plants for data centers).
Here: constructive on oil in the $60s, not underwriting $100; bullish gas and a narrowing Marcellus/Haynesville basis to Henry Hub → WHK (30:48).
Watch for

44:56 9. In a speculative sector, own the monster or the metal

The repeatable method
  1. Confirm the commodity thesis is present-tense (a shortage from existing demand, not a hypothetical future).
  2. Screen out the sector's quality problem: skip small caps and development stories.
  3. Choose between the largest strong-jurisdiction producer with operating leverage and an embedded special situation, or the physical commodity trust.
Here: uranium → CCJ (plus Westinghouse optionality) or physical SRUUF (SPUT), which he owns personally.
Watch for

Methods distilled from the public YouTube video (Market Talk with George Noble, 2026-SEP-12) for personal study. Not investment advice.