Actionable insights — Rising Bond Yields Are EVERYONE'S Problem

14:31 1. Strip mark-to-market and other one-time income before you trust any PE

1. For any company that holds equity stakes in fast-rising private businesses, find the "other income" line. Unrealized appreciation on those stakes is booked as earnings even though no cash changed hands.
2. Compute the ratio: one-time markup ÷ reported earnings. Anything above a small fraction means the headline number is telling you about a valuation, not about a business.
3. Do it at the index level too: take the reported index earnings-growth rate and subtract the points attributable to mark-to-market accounting. What's left is the organic rate.
4. Re-run the multiple on the organic number before comparing it to history. And remember the symmetry: whatever inflated earnings on the way up produces write-downs on the way down.

• "Other income" as a share of pre-tax profit; new private-round valuations that will be marked next quarter; the first down-round, which reverses the whole mechanism.

16:57 2. Check the cash, not the earnings — and check how few names carry it

1. When an index is described as "flush with cash" or "record margins," decompose it: how many companies actually produce the aggregate? Aggregates hide concentration.
2. Then test those specific names on free cash flow, not net income. Ask the operator's three questions the accounting can't fake: can it pay salaries, fund capex, and service interest?
3. If the handful of names carrying the aggregate have flipped to negative free cash flow, the index-level statistic is stale — and the market is quoting a balance-sheet strength that no longer exists.

• Free cash flow turning negative at the index's largest contributors; capex guidance rising while operating cash flow flattens.

18:49 3. Read the credit market for the discernment equities don't have

1. When you suspect a theme is deteriorating but the stocks won't confirm, go to the debt of the same issuers: the yield spread over Treasuries and the cost of default protection (CDS).
2. Credit differentiates issuer-by-issuer well before equity does, because lenders get no upside and are paid only to price downside. Widening spreads inside a single theme = the first real discernment.
3. Use the divergence as a clock, not a trigger: while equities are still making records, the credit signal tells you which way the resolution goes, not when.

• Hyperscaler bond spreads vs the index; CDS on the biggest AI borrowers; the gap between the credit signal and the equity price as the setup.

20:44 4. Build the issuance calendar — forecast rates from the supply of paper

1. Add up who has to sell paper in the next twelve months, in this order: government bills that auto-roll, government notes and bonds, corporate maturities to refinance, municipal and agency needs, and finally new corporate borrowing from any capex boom.
2. Compare the total against the pool of capital willing to buy it. The question is never "can the government fund itself?" in isolation — it's "who wins the auction for a finite pool?"
3. Ground it in a real data source rather than headlines: SIFMA publishes quarterly bond-issuance data by sector. Ratio to watch: private-sector issuance ÷ federal issuance.
4. Invert the traditional crowding-out story. If private issuance rivals the government's, corporates are crowding out the Treasury, and the long end — the benchmark everything else prices off — pays the price. Higher-for-longer follows mechanically, no Fed forecast required.
5. Corollary for allocation: the more bullish you are on the capex theme, the more issuance it implies, and therefore the more committed you should be to higher rates. Bullish-on-AI and bullish-on-bonds is an inconsistent pair.

• The private/federal issuance ratio each quarter; the bill share of Treasury funding; foreign official selling (Japan, China) as the marginal seller on the long end.

22:25 5. Track the buyback → net-issuance flip as a regime signal

1. Identify the market's structurally largest buyer. For a decade-plus it was corporates themselves via buybacks — often the single biggest purchaser of equities.
2. Net it out properly: gross buybacks minus new equity issued. The sign of that number, not its size, is the regime.
3. When it flips negative — companies as net issuers — the biggest bid has become an offer, and the marginal buyer must be replaced (here, by retail). Treat it as a slow structural headwind, not a timing signal; it works quietly.

• Quarterly net buyback data turning negative; secondary offerings and mega-IPOs clustering; retail flows being cited as the replacement bid.

31:08 6. Separate the cyclical case from the secular one before forecasting

1. Sort your reasons for a view into cyclical (issuance, oil, a capex boom — things that mean-revert) and secular (trade architecture, reserve-currency demand, demographics — things that take decades).
2. Anchor the secular leg on a historical analogue and run it backwards: what caused the last 40-year trend, and is that cause reversing? Globalization from the fall of the Berlin Wall pushed rates lower from 1980; deglobalization should beget the reverse.
3. Add the second-order effect: a multipolar world is less reliant on the dollar and therefore on Treasuries (less buying, sometimes outright selling), and reshoring raises labor costs — both inflationary, both rate-positive.
4. Use the split to size and to pace. A cyclical view justifies a trade; a secular view justifies a portfolio, held for years, entered early.

• Foreign official Treasury holdings; reshoring announcements and their labor-cost math; any policy that raises the cost of goods for national-security reasons.

32:27 7. The tub: decide asset-class leadership from where the marginal dollar goes

1. Model liquidity as a tub filling each day. The real economy takes what it needs to move a dollar of output; the overflow goes into financial assets.
2. Ask which way the cost of moving that dollar is trending. Falling inflation and rates ⇒ the economy needs less, overflow grows, financial assets lead (the 1980–2020 experience).
3. Rising costs of goods, labor and capital ⇒ the economy siphons more of the marginal liquidity, the overflow shrinks, and real assets lead. The relative call falls out of the direction of the cost of output, not from an earnings forecast.
4. Cross-check it with the supply side: financial claims can be printed without limit ("hyper scale" issuance), hard assets can't be manufactured out of thin air. Scarce supply against infinite supply is the relative-value trade.
5. Confirm with a long-horizon relative chart (e.g. commodities vs the S&P) sitting at secular lows before you position — and expect the turn to take years.

• The direction of unit costs (goods, labor, capital); the commodities/S&P relative line turning up off secular lows; issuance volumes as the paper-supply side of the ratio.

30:18 8. Price policymakers by their constraints, not their rhetoric

1. Write down what the official campaigned on, then write down the arithmetic they inherited. Where the two conflict, the arithmetic wins.
2. Check the revealed action rather than the speech: is the balance sheet actually shrinking? Is the funding mix actually shifting away from bills?
3. Ignore the instrument that doesn't work. If the last several policy-rate moves saw the long end go the other way, stop trading the policy rate and watch the balance sheet — the only tool with real leverage over a market this levered.
4. Then ask what would force the reversal, and how big it has to be: here, a credit crisis large enough to make the Fed the buyer of last resort for Treasuries.

• Weekly balance-sheet size vs the stated plan; the bill share of issuance; deficit trajectory during a growth boom — if it isn't shrinking now, the grow-out-of-it plan has failed.

37:54 9. Stress-test long-dated liabilities at the top, not at the bottom

1. Look at any pool of long-dated promises (pensions, endowments, insurers) and read its funding gap in the context of where markets are. A shrinking deficit achieved only because assets are at record highs is not an improvement — it's the best case already spent.
2. Inspect what closed the gap. Reaching for private equity and private credit to escape a zero-rate world adds valuation risk that isn't marked daily.
3. Sanity-check those marks against observable transactions — secondary sales at 50 cents on the dollar or less tell you the carrying values are fiction.
4. Then follow the political consequence to its fiscal end: the shortfall doesn't disappear, it gets socialized. Size the resulting deficit and central-bank balance sheet, and hold the assets that benefit from that outcome.

• Funded-ratio reports dated at market peaks; secondary-market marks on PE/private credit; the first large public plan seeking legislative relief.

59:07 10. Be early, own the boom through its inputs, keep dry powder in bills

1. For a decade-scale secular view, build a portfolio you don't have to "monkey around with" and accept being early: "I would rather be early to this than late." Position before the trend is confirmed, then sit.
2. Don't fight the boom you disbelieve — own its physical inputs instead. If you're an AI bull you have to be an energy and copper bull, so exposure to the energy complex captures the capex cycle and keeps you in protected hard assets.
3. Hold a partially-filled position in the asset that hasn't found its footing yet (gold), rather than waiting for a signal that only arrives after the move.
4. Keep the rest as dry powder in T-bills — paid to wait, with no forced view.
5. Pre-commit the condition for the next trade rather than the price: she will buy Treasuries only after seeing "a little bit more agita," i.e. after the stress shows up, not before.

• Gold basing/regaining its footing; energy's share of the capex bill; the "agita" (credit stress, forced selling) that would make long Treasuries a buy.