Actionable insights — The 2029 Oil Shock Investors Aren't Ready For

3:42 1. Refuse the unknowable horizon; answer the knowable one

1. Split any commodity question into two horizons: the near term (driven by events nobody can forecast) and the out-years (driven by physical investment already made or skipped).
2. Decline the near term explicitly, and say why: "neither I nor the Ayatollah nor Ayatollah Netanyahu or Ayatollah Trump has any sense of what the near-term oil price is going to be. And I don't spend much time trying to figure out something that it's impossible to know the answer to."
3. Separate the artificial shortage from the structural one. An artificial shortage is a political event and "could end with an armistice"; a structural shortage is arithmetic and "can't be ended by an armistice."
4. Pick a date far enough out that the physical facts dominate the headlines — here 2029–30 — and build the position for that date rather than for the tape.
5. Accept the interim: "it wouldn't surprise me… to see the oil price decline a lot. But I'm okay with that. I'm not a trader."

• Sustaining-capital disclosures and rig counts (physical facts) rather than ceasefire headlines.
• Your own portfolio containing positions justified only by an event you admit you can't forecast.

5:03 2. Score a resource company on sustaining capex, and treat a fat dividend as a warning

1. For each producer, ask where the operating cash flow went: into sustaining capital and new projects, or into dividends and buybacks?
2. Label the second pattern honestly — a company "diverting money from new project investment and sustaining capital investment [is] cannibalizing itself." It is a liquidation dressed as a yield.
3. Invert the market's preference. Because "the market seems to favour cannibalization," the spenders are cheaper: "many" US companies do make the investments, "they're just not as popular as those that don't."
4. Apply the forward test to the dividend itself: "if you defer your sustaining capital investments, you're not able to pay the dividends four or five years out." A stingy dividend today is the one that still exists later.
5. Allow one exception, and require it to be earned operationally: a company can under-spend and still grow if its drill-bit efficiency has genuinely improved — verify by whether reserves and production are both rising.

• Capex guidance cut while the dividend is raised; reserve-replacement ratios below 100%.
• For the exception case, whether efficiency gains are still expanding or have plateaued.

6:30 3. Invert the "who's most exposed" question using the decline curve

1. When a shortage thesis names the obvious victims, check the production physics rather than the politics.
2. For each producing region, ask how fast output falls if drilling stops. Shale is front-loaded: "you get a substantial part of the net present value of the production of that well out in the first 18 months," so output "falls even more precipitously in the shale basins than it does in conventional reservoirs."
3. Conclude accordingly, even when it contradicts the consensus: "the idea that the deferral of sustaining capital investments is going to impact Brazil and Iran and Saudi more than it's going to impact the United States and Canada is wrong."
4. Translate the finding into a screen: in a steep-decline basin, a producer's continued drilling budget is not growth spending, it is survival spending. Fund the ones still doing it.

• Base-decline rates disclosed in operators' presentations; recompletion schedules as well as new drilling.
• Any producer whose flat production is being maintained by drawing down an inventory of drilled-but-uncompleted wells.

12:46 4. The cage signal — trade the ticket ratio, not the chart

1. Find a countable proxy for what real people are actually doing, not what commentators say. His was the trade-processing room: the ratio of buy tickets to sell tickets.
2. Set a hard threshold at a genuine extreme — 5:1 — so ordinary noise never triggers it.
3. Act mechanically and modestly in the opposite direction: "any day where there was a 5:1 preponderance of buy tickets to sell tickets, the next day I'd sell something… any day that there was a 5:1 preponderance of sell tickets over buy tickets, I'd buy something."
4. Keep it as the only technical input, and be honest that it is: "these extremes in sentiment are probably the only indicator that I've ever learned how to use."

• A modern proxy you can actually observe: fund-flow extremes into a sector ETF, put/call ratios, or retail-platform net-buy data.
• The discipline of trimming something rather than everything — the signal sizes an action, not a regime change.

13:43 5. Hockey-stick symmetry — sell the parabola, buy the explainable crash

1. Treat a hyperbolic advance as a sell trigger on its own terms: "the backside of a hockey stick is just as steep as the front, but it's a lot less fun if you're long."
2. Recognise the psychology that creates it — "there's something about something becoming rapidly more expensive that to an odd class of people makes it more attractive" — and act before the crowd it attracts arrives.
3. Run the mirror image on the downside, but only where you can explain the decline: a 15–20% drop caused by a fund meeting redemptions "had nothing to do with the company, had nothing to do with the broad market."
4. Verify the cause before buying — thin bids plus a forced seller is a gift; thin bids plus deteriorating fundamentals is a warning.

• Fund redemption/gate announcements, tax-loss season, and index-deletion dates — the calendar of forced selling.
• Volume spikes with no company news; a price move that no filing explains.

19:37 6. Get in the way of the money — front-run index inclusion on a two-to-three-year lag

1. Accept that passive flows are not a market you can fade: "I think you have to use it. I think you need to get in the way of the money."
2. Understand what index construction actually rewards: "trading liquidity, simple market cap and heft" — size itself becomes a valuable attribute, independent of business quality.
3. Look for wide valuation gaps between small and large companies in the same sector. There are only two outcomes: "either the value arbitrageurs arb that away or the big companies take over the little companies. End of discussion."
4. Buy the small company and give the process time — "front running the ETFs, if you give yourself a two-year or three-year lag time, is really truly simple."
5. Know where the re-rating comes from: "a single asset producer in the mining business is viewed as being riskier, which it is, than a multi-asset producer. That discount goes away when the single asset producer gets taken over."

• Single-asset producers trading at a large discount to multi-asset peers on the same metric.
• Index-provider size and liquidity thresholds, and which companies are approaching them.

20:25 7. Sort M&A into strategic and tactical — they need different screens

1. Strategic: find deposits worth more to a specific neighbour than to their own shareholders. The test is infrastructure — a deposit too small to justify its own mill is nearly worthless standalone, but is simply extra ore to a major "that has a mill within trucking distance and doesn't have to build a $350 million mill."
2. So screen geographically: map orphaned deposits against existing mills, smelters, ports and pipelines, and identify the one logical buyer.
3. Tactical: find managements deliberately assembling scale with no operating synergy at all — "you combine companies that have absolutely no operating synergy… In other words, you construct a new major."
4. Score those on the index test instead: production run-rate, market cap, liquidity and pipeline — whatever the indexes require. The payoff is inclusion and the perpetual passive bid that follows.
5. Judge probabilities, not certainties: "understanding the likely — not the certain, but the probable — outcome, and tailoring part of your investment thesis to getting in the way of the money."

• Mill capacity utilisation at nearby majors — spare capacity is acquisition appetite.
• Merger rationales that cite "scale," "re-rating" or "index eligibility" rather than cost synergies; those are tactical by definition.

18:35 8. Own the manager, not the product — "would you rather pay or be paid?"

1. When you are about to hold several funds from one sponsor, count them and count the fees you'd pay.
2. Check whether the sponsor is listed. If so, compare: buying the manager gives "an indirect interest in all 60 of them" plus a dividend, versus buying four or five products and paying a management fee on each.
3. Ask the question in one line: "would you rather pay or be paid?"
4. Price the difference in risk honestly — the manager is an operating business whose earnings track fund inflows and asset values, so it is more volatile than the products it sponsors. He calls it "the added risk."
5. Disclose conflicts before using the argument on anyone else; he names his own immediately.

• Assets under management and net flows at the manager — the real earnings driver.
• Fee compression and whether the sponsor's products are differentiated (physical trusts) or commodity-like (index funds).

10:58 9. Hold both halves: markets work, and inevitable isn't imminent

1. Start from the first principle: "the cure for high prices is always high prices. And the cure for low prices is always low prices." High prices summon supply and destroy demand; low prices do the reverse.
2. Immediately attach the second: "what is inevitable is not necessarily imminent. Although it has to correct, it doesn't have to correct quickly."
3. Use the pair to size and to schedule — conviction from the first, patience and position size from the second. Holding only the first is what bankrupts people who are eventually proved right.
4. Audit any thesis you hold against both: can you name the mechanism that forces the correction, and have you allowed enough time for it?

• The supply response your own thesis implies — if high prices haven't yet summoned new supply, the cycle is not over.
• Positions sized as though the inevitable were imminent (leverage, options, short-dated expressions).

42:05 10. Find the popularly believed fallacy — then check it against a long arithmetic record

1. Identify the belief that is driving capital allocation rather than the belief that is loudest. Here: that peak oil demand arrives in 2030, so maintaining production is pointless.
2. Trace who holds it and what they control — "institutional investors to some extent control the purse strings and I think the major banks have had a role to play."
3. Test it against a long, boring measurement rather than a forecast: "$10 trillion over 45 years in alternative energies… has reduced the market share of fossil fuels" from 83% to 81%.
4. Add the demand nobody models: "this isn't all about data centers. This is about making the poorest of the poor less poor… A billion people on Earth have no access to primary electricity."
5. State the trade plainly and take the other side: "I think that supposition is wrong. I think it's easy to bet against."

• Market-share data (not spending data) for the substitute; spending proves commitment, share proves displacement.
• The moment institutional mandates change — that is when the discount closes, and it closes fast.

44:55 11. Follow the retreating lender — capital withdrawal is both the cause and the confirmation

1. Track where credit is leaving, not just where equity is cheap. Banks exiting a sector is a harder signal than sentiment because it is contractual and slow to reverse.
2. Read it two ways at once: as confirmation of the supply thesis (less capital now means less production later) and as a business opportunity (less competition for the loans that remain).
3. Check whether the retreat is credit-driven or narrative-driven. If the borrowers are creditworthy and the lenders are leaving for reputational reasons, the pricing is mispriced in your favour.
4. Prefer the position where you are paid to provide the scarce thing — capital — rather than competing for the scarce asset.

• Bank sector-exposure disclosures and net-zero financing pledges; reserve-based lending availability for mid-cap E&Ps.
• Borrowing spreads for energy credits versus comparably rated industrials — the size of the narrative premium.

27:21 12. Price permitting delay as an NPV killer, not an inconvenience

1. Before valuing any undeveloped deposit, ask when production actually starts — then discount from that date, not from today.
2. Use the rule of thumb: at an 8% discount rate, "an 11-year delay in the start date takes away all of the net present value, 100% of it."
3. Weight the jurisdiction's permitting record more heavily than the geology when the timeline is open-ended. Resolution has "well over a billion tons, average grade 1 and a half percent, more than three times the average grade of mined copper in the world," highway, rail, power, water and a mining town next door — and has been permitting for 28 years.
4. Separate the policy fix from the subsidy: subsidies don't shorten timelines. "They just need to get out of my way."
5. Note what a genuine green light would look like — if the majors believed they could be permitted, "they'd be there like white on rice." Absence of major-company effort is itself evidence about the permitting odds.

• Permit-stage milestones with dates, and the litigation history of the jurisdiction rather than the project.
• Whether a deposit's owner is a major or a promoter — majors don't spend where they don't expect permits.