Everything Is a Bet on "When"

Happy Sunday.

Government bond yields hit multi year or multi decade highs in five countries this week. Diesel set an all time record on Friday, $5.85 a gallon. August payrolls came in at 162,000, three times what anyone expected, and the odds of a Fed hike in eleven days moved to sixty percent.

Every headline you read used the word rout. Yep, basically a messy, unstable, and frustrating to navigate market.

But thesame week, American corporate pension funds posted their best year since 2007.

Sixteen straight years underfunded. They bottomed near 77 cents on the dollar in 2012, and for most of the last decade the line in every annual report was some version of we are working on it. S&P 500 plans finished last year at 103 percent funded. Milliman has the hundred largest at 112 percent as of July. A few companies are far enough past the line that they are selling the obligations to life insurers and walking away from the problem.

They did not earn it.

No stock got picked. No trade got made. A number in a spreadsheet moved and a problem that outlasted three Fed chairs went away, and it was the same number everyone else spent the week calling a crisis.

TLDR: Interest rates went up, so a dollar arriving years from now is worth less today. Pensions won big on that —> the retirement checks they owe decades out suddenly cost less to cover. You lose on it, because your 401(k) and the whole AI buildout are bets that only pay off far in the future, and the market just marked those down.

Real rates did all of it

This is the part I think most people have backwards, and Franklin Templeton put the number on it this week.

Inflation expectations drawn from TIPS have barely moved this year. Real rates did effectively all of the work in yields. In plain English, the bond market is not forecasting runaway prices. It is charging more for time.

That is what a real interest rate is. A price on waiting.

And everything in a portfolio is a claim on some future date. A stock is a claim on 2035 earnings. A house is a thirty year claim you financed at 6.57 percent. A data center is a claim on revenue nobody has invoiced. A 401(k) is a claim your fifty year old self writes to your sixty seven year old self. Raise the price of waiting and you have re-sorted the whole board by how far off each payday sits. Bond people call that distance duration, and this year it is the only sort the market has been running.

In 1928 a Cambridge mathematician named Frank Ramsey wrote the paper that gave economics the discount rate (Ramsey formula). He also wrote that valuing today's enjoyment above tomorrow's was ethically indefensible and came from a weakness of the imagination. He died two years later at 26. What his number measures is impatience, and the price of it just hit a two decade high.

Who stopped buying

Mohamed El-Erian put it in one sentence this week, and it had nothing to do with inflation or the Fed.

The reliable buyers are going away. China is less willing for geopolitical reasons. Japan and the Gulf states have problems at home. The Norwegian sovereign wealth fund is rethinking its allocation to U.S. government bonds, and the size is small but the signal is not. Against that, he said, the issuance coming from governments, hyperscalers and ordinary corporates far exceeds what you can count on in reliable demand. His read was a fundamental imbalance, more than inflation or Fed credibility.

The buyers who leave are the patient ones. Foreign central banks, sovereign funds, Japanese life insurers, the accounts that bought thirty year paper and held it for thirty years. They supplied the world's willingness to wait, and they are stepping back while the demand for waiting goes up.

Gross U.S. debt passed $40 trillion last month. Total public and private borrowing worldwide runs near $350 trillion, about three times what the world produces in a year. Every dollar of it asks somebody to wait.

The tape shows the shortage. The U.S. 10-year closed the week at 4.78 percent, the highest since November 2023. Japan's 10-year touched 3 percent for the first time since 1996. Gilts hit a post-2008 high, bunds went back to 2011 levels, and France has replaced Italy as the eurozone's problem.

The war is what moved the price this week. Brent closed near $96, diesel set its record, Friday's payrolls came in at 162,000, and the odds of a hike on the sixteenth went to sixty percent. UBS now looks for two this year. Six months ago the argument was how many cuts. But the oil shock explains the move, not the level. Take it away and the buyer problem is still sitting there.

Today’s Sponsor

Invest Alongside a Manager with a 40-Year Track Record

Private real estate has long been a cornerstone of institutional portfolios, but individual investors have often had limited access to the asset class - or have had to invest through pooled funds and intermediaries.

Lightstone DIRECT offers a different approach. Accredited investors can invest directly in select real estate opportunities alongside Lightstone, a vertically integrated owner/operator with more than four decades of experience and a $12 billion real estate portfolio.

We invest 20% or more of the equity in every offering, aligning our interests with yours from day one. Our in-house team oversees acquisitions, asset management, and investment execution, with property operations managed by Lightstone and, where appropriate, experienced third-party operators.

No blind funds. No intermediaries. Just direct partnership.

Accredited Investors only. $100k minimum investment.

Explore Opportunities

^{This communication is for informational purposes only and does not constitute an offer to sell or a solicitation of an offer to buy any securities. Any such offer or solicitation will be made exclusively through the definitive offering documents. All investments involve risk of loss, including the potential loss of principal. Past performance is not a guarantee of future results. Any targeted returns or projections are forward-looking statements, are based on current assumptions, and are not guarantees of future performance. Actual results may differ materially.}

Where the AI buildout sits on the curve

I have thrown the eleven trillion dollar buildout number at you before and I will not rerun it. Look at the structure instead of the size. Roughly seven trillion of it is debt financed, and the revenue arrives years after the concrete does. That is a long dated asset funded in the tightest market for patience in twenty years.

Nvidia $NVDA put $3.5 billion into MediaTek this week, bought Hugging Face for around $13 billion, and backed a $35 billion cloud deal between Anthropic and Lambda. According to people familiar with the deal, Nvidia itself holds the lease on the data center.

When the market will not hold a long dated claim at a price the borrower likes, the vendor holds it instead. The duration did not go anywhere. It moved onto the balance sheet of the company selling the chips.

Then there is what Kevin Warsh did at Jackson Hole. El-Erian's read was that the least covered thing Warsh said was the most important. He called AI a potential factor of production, a real supply side force rather than a story.

So the thirty year yield and the AI trade are now the same argument in two units. One says the future is worth less today. The other says the future is worth an enormous amount, starting soon. Both cannot hold, and the bond market gets to be right first, because debt matures on a schedule and productivity does not. Whatever the buildout eventually produces, the 2027 and 2028 maturities arrive on dates already printed.

Workers stopped switching jobs

The same behavior showed up this week somewhere that has nothing to do with bonds.

Americans have stopped changing jobs. The separation rate sits near its lowest level since the years after the financial crisis. Low hire, low fire. Nearly two million Americans have been looking for work for six months or more, and the long term unemployed are 27 percent of the total, up from 21 percent two years ago. Only 33 percent of people looking think it is a good time to find a quality job.

A 42 year old technical writer in Irvine saw what she called her dream job posted last month and did not apply. Four years of seniority, a boss she likes, and a feed full of freshly laid off friends. She taught herself AI coding tools on her own time instead, made herself harder to replace, and got promoted.

A career is a claim on a future date too, and job hopping is the long dated version of it. When waiting costs more and the payoff looks less certain, people stop trading the position and take the sure money.

Betsey Stevenson at Michigan describes the result as a crowded restaurant where every table filled ten minutes ago. It does not matter how many tables there are. Nobody gets seated until somebody stands up, and seeing the line, nobody stands up.

Ugly Truth

The case against all of this is strong and it comes from serious people.

Barings says position defensively in short duration. Franklin Templeton, the same week, says extend duration along the U.S. curve. Two credible shops, opposite calls, same data.

I went looking for which one was wrong and I do not think either is. Franklin writes for money with long liabilities, so it can afford to be paid to wait. Barings runs funds that get marked every quarter, so it cannot. Same data, different clock. Which either confirms everything above or means I found my own thesis in a coin flip.

Franklin's larger point is the one I have the least answer to. If real rates are rising because trend growth is speeding up, because productivity is finally improving and returns on capital and margins sit at records, then the higher rate is what an economy that works better costs. Warsh's factor of production line points the same way. In that version, this letter is careful research in the service of worrying.

The tape half agrees. The VIX sat around 15 all week with a war on, diesel at a record and a live hike on the board. The S&P is up 13 percent on the year. Greg Abel looked at multi decade high Japanese yields and said they are not a problem for the trading houses right now. Deutsche Bank figures the 10-year would have to reach about 5.5 percent over the next year, or 6.4 percent over two, before the coupon stops covering the price decline. That cushion did not exist for fifteen years.

One chart keeps pulling at me. The U.S. 10-year spread over most developed peers sits well above the top of its historical range, near record wides against Switzerland and China. Either the world has correctly decided America's fiscal path is worse than everyone else's, or the rest of the world is too cheap and this closes hard. Spreads that wide have not historically closed gently.

Today’s Sponsor

Farmers need alternatives to chemical-intensive farming. Greenfield Robotics is already operating in commercial fields: 82 robots, 16 states. Reserve now for 10% bonus shares when you invest.

See Investment Details

What a household does not mark

The usual reminder, I am not a financial advisor and none of this is advice.

A pension fund is 112% funded today because it keeps both sides of a balance sheet.

Assets and liabilities, both marked, both discounted at the same rate. Rates went up, the liabilities fell faster than the assets, and the plan won without anyone lifting a finger.

You have liabilities too. Retirement spending, tuition, the mortgage, care costs for a parent. Nobody marks them. So the same move that rescued the pension reaches a household as a worse mortgage quote and a bad month in the bond sleeve, with none of the offset showing up anywhere you can see it.

Fidelity says the average 401(k) balance in your thirties is $75,200, and $156,800 in your forties. That participant holds the longest dated book in the country and has never once been told what a discount rate is.

Two questions, then.

When do I need each dollar? A pension knows the year it owes the money and buys to match. Most households pick an allocation and never write the dates down. If you cannot name the year, you are guessing at the one thing that now prices everything else.

And am I selling patience or buying it? A retiree laddering five year paper at a real yield is selling it into a market that is short. A borrower rolling debt in 2027 is buying it at the worst price in twenty years. Same economy, opposite sides, and which side you land on has almost nothing to do with how smart you are.

Time was free for fifteen years, so nobody priced it. Now it has a price, and everything you own is a bet on when.

Stay curious 😎

- John