I read all seventeen pages of the Federal Reserve's projection tables and the full press conference transcript this week so you would not have to, which is either dedication or a cry for help. I still cannot tell you where oil trades on Friday. I can tell you why the stock market went up on a rate hike, and why that was the right answer.
Here is the whole letter in one sentence: the Fed raised the price of money, and the companies at the top of the S&P 500 are the ones selling it.
What happened this week
The Fed raised rates a quarter point on Wednesday, its first increase since the summer of 2023, twelve votes to nothing. Europe's central bank raised the week before. Japan raised on Thursday. The last time central banks moved together like that they were chasing the inflation of 2022.
The reason is at the gas station. Diesel is over six dollars a gallon and the world price of oil closed above $130 a barrel on Tuesday against the low sixties a year ago. The Strait of Hormuz, the narrow exit from the Persian Gulf that a fifth of the world's oil sails through, has been mostly shut since the war with Iran began in March. Washington fought the price with the Strategic Petroleum Reserve, which is exactly what it sounds like, a very large government stash of oil in salt caverns under Texas and Louisiana. That is why oil fell to the eighties over the summer and nobody panicked. The stash is now a third emptier than in March, the promised refill has not started, and the ceasefire collapsed in July.
The Fed's own chair was asked the obvious question on Wednesday, that a quarter point hike does not reopen the strait. His answer was that the Fed cannot fix any one price, but it can make sure the oil price does not spread into every other price. So the hike is insurance, and he said something else that matters more: he was "hard-pressed to describe broad financial conditions as restrictive," and the committee "removed a dose of accommodation." That is a central bank tapping the brakes on a car it thinks is running well, with core inflation already near its target, and a tap is a very different thing from a slam.
Why the market is right about the index
Stocks went up on the hike, and the commentary called that irrational. It was not. Three things explain it.
First, earnings. The companies in the S&P 500 are on track to grow profits by about a third this year and another fifteen percent next year, and every one of the eleven sectors is growing. Since June the index price rose about two percent while expected earnings rose about nine, so investors are paying less per dollar of profit than they were in the spring. The oil shock is part of that: the energy sector's profits roughly doubled and refiners' quadrupled, so the thing hurting you at the pump is helping the index.
Second, the biggest companies do not borrow. When rates go up, a company with a pile of cash earns more on it and a company with a pile of debt pays more on it. Apple has about sixty billion dollars more cash than debt and issued no bonds this year. Nvidia and Alphabet are net lenders too. The stock market is weighted by size, so the index is dominated by companies for whom a rate hike is a raise. The market did not learn some lazy lesson about buying rate hikes. It did the math on who pays the interest.
Third, politics. The midterms are six weeks away, on November 3, and the 2028 race starts the morning after. Nothing moves a president toward a deal with Iran faster than four dollar gasoline in October. Oil already fell below a hundred dollars on Monday on talk of talks, and I expect that pressure to keep pushing the price down through the fall, whatever the tankers are doing. Tom Lee at Fundstrat, who has been right about this market for three years, called Wednesday's hike "max hawkishness" and a buying setup, and sees the index above 8,200 by year end. Ed Yardeni, another long-time bull, cut his target to 7,900 the same day and said the quiet part: earnings will be fantastic, but a five percent bond yield takes something off what people will pay for them. I think both are describing the same market. It goes up, and it goes up more slowly than the earnings.
Where the damage actually lands
The Fed's projections show no rate cuts next year, and the 2026 number implies one more hike before Christmas. They also show inflation falling to 2.3 percent next year, which you can only forecast if you assume oil comes down. So the Fed is already assuming the market's happy oil scenario, and it still plans to sit at four percent for two years.
A two-year plateau does not hurt Apple. It hurts whoever has to refinance. Look at our own screener. Over the past year the typical company with little debt returned about thirteen percent, the typical company carrying more than three years of earnings in debt returned three, and most of the names that lost money sit on the indebted side of the chart. Size has nothing to do with it, because small caps as a group have beaten the S&P this year. Debt is the whole story.
The bond market says the same thing, if you look at the right tier. The headline junk-bond spread, the extra interest risky companies pay above the government, got cheaper this year, which is why people call credit complacent. But the riskiest tier, the companies rated CCC, got about two hundred basis points more expensive and hit its widest point of the year the day before the hike. The smoke detector works. It is just in the basement.
Picture a landlord who owns the whole building outright and a tenant on the top floor with three credit cards. The Fed just raised interest rates. The landlord's savings account pays more, so he sleeps fine. The tenant's minimum payment went up, so he is selling his couch. From the street the building looks the same. That is the S&P 500 right now: the index is the landlord, the average small indebted company is the tenant, and a two-year plateau is two years of minimum payments.
One honest complication inside the giants. The AI build is now partly funded with debt. Amazon is a net borrower with a fifty billion dollar quarterly capital budget, and Alphabet, Meta and Nvidia all sold bonds this year, which was unheard of a few years ago. The cash-rich label fits Apple, Nvidia and Alphabet cleanly and Amazon much less so. Inside the seven, the same rule applies: own the lenders.
Where things go from here
Stocks. The index keeps going up, and it does not go in a straight line. History says the year after a first hike is choppy, up modestly, with a double-digit pullback somewhere in the middle, and Yardeni's math on the bond yield says the same. I think we see 8,000 by the end of the year, against 7,650 today, and I would treat any drop on the way there as a gift.
Earnings keep compounding, the multiple holds, and the fourth quarter does what fourth quarters usually do. Up about five percent.
Oil. The government's forecasters have Brent near $74 next year. Political pressure says it heads that way sooner than the tankers justify, and I expect the world price to average somewhere between $80 and $95 through June, well below Tuesday's spike. The reserve is still empty and the strait is still impaired, so any deal that wobbles sends it right back up.
Who wins inside the S&P. The lenders. Apple, Nvidia and Alphabet among the giants. Visa, Mastercard, ADP and Progressive from our screener's low-debt list, all near or below ALAN's fair value. The gold miners Newmont and Agnico Eagle, which have gone nowhere while gold sat flat and are the cheapest insurance against a war that does not end. And the energy producers with clean balance sheets, EOG, Suncor and Canadian Natural, as a hedge only, because they have already been the trade of the year and their profits are forecast to fall next year even if oil stays high.
Who loses. The tenants. Small companies with more than three years of earnings in debt, unprofitable growth stocks that need a rate cut to justify their price, indebted real estate, and long-term bond funds, which are already down six percent this year. Junk bonds rated CCC. Amazon relative to its peers, until the borrowing stops.
Safe havens. Three-month Treasury bills or a money market fund that holds them pay four percent with no risk to your principal. Gold has not worked this year, which surprised me, and I would own the miners over the metal.
What I am doing with my own money
I am not going to tell you what to do. I will tell you what I own, and you can see it any day on the site, percentages only. My accountant and I have an understanding about the dollar figure.
- Nvidia40.9%
- CrowdStrike29.9%
- Tech index fund (VGT)11.7%
- AMD6.6%
- Microsoft3.6%
- Meta2.0%
- Apple1.7%
- Alphabet1.4%
- Mastercard1.0%
- Netflix0.9%
- Visa0.4%
Seventy percent of my portfolio is two companies, Nvidia and CrowdStrike, and I am aware that is not what a textbook looks like. Another tenth is a plain technology index fund, which is the closest thing I own to a seatbelt. Nvidia is the cleanest lender in the group and the model still reads it below fair value. CrowdStrike is on the other list, the one our model calls expensive, and I hold it anyway because I think security spending is the last line item a company cuts, and because I bought it at less than half of today's price. The rest of the book is AMD and the lenders: Microsoft, Meta, Apple, Alphabet, Mastercard, Visa. Nothing in it needs a rate cut.
So what am I doing about everything above? Nothing. I am not a trader. I buy what I think are the highest quality companies in the world and I hold them for years, through weeks like this one, and by my own accounting that has returned about 36 percent a year since January 2019. If I were a trader I might look at an energy hedge, or at the pullback trade, and the letters exist so you can decide those things for yourself. The book stays put. I own no small caps, no junk bonds, no long-term bond funds and no real estate, and nothing in this week's news makes me want any of them.
Two dates I am watching, with a coffee rather than a sell button: November 3 for the midterms and what they do to the Iran talks, and December 16 for the Fed's next projections.
What would make me wrong
The index closes the year below where it is today. That would mean the oil tax finally broke the consumer, or the debt-funded AI build hit a credit wall, and the landlord's building lost value after all.
The riskiest junk spread narrows back inside seven hundred basis points. That would mean the tenants found cheap money again, the plateau was a bluff, and the whole dispersion story was noise.
The small, heavy-debt group in our screener starts beating the index. Then I will write that here by name. But I will tell you this: if it does happen, I will be shocked, and I will owe my buddy at least $100.
The Fed raised the price of money. Most of the S&P 500 is the one selling it.
Fair values, price targets, ranges and the cheap or expensive readings derived from them are ALAN's own model outputs, computed from public filings, market data and analyst estimates that may be incomplete, delayed or wrong. They are not investment advice, not a recommendation or solicitation to buy, sell or hold any security, and not a prediction of where a price will go. ALAN is software, not a registered investment adviser or broker-dealer. Past model accuracy does not predict future accuracy. Do your own research and consider consulting a qualified financial adviser before making any investment decision.