Put Down the Phone

September 26, 2026

The wall of worry has never been taller, and I think most of it is made of noise. Every bull market climbs one, and the adage is old enough that nobody remembers who said it first. What has changed is how the wall gets built. A generation ago the worries arrived once a day on the front page and once a night on the evening news. Today they arrive every few minutes on a phone, sorted by software that has learned alarm keeps us looking, and the habit even has a name: doom scrolling. This week the scroll ran from Iran to the midterms to Kevin Warsh, and the S&P 500 has now fallen on the day of each of his first three meetings as Chair. None of it tells me what the companies in the S&P 500 will earn next year. The signal is corporate earnings, growing at a pace usually seen only coming out of a recession, and a world economy whose purchasing managers keep reporting faster growth. Two weeks ago I let a slowdown in the economic data pull my attention away from that signal. This letter is about separating the two, and the chart below is where I start.

S&P 500 forward twelve-month earnings estimates, 2016 to 2026. Source: Bespoke Investment Group.

Market Fundamentals

Earnings are the signal, and they haven’t wavered. Bespoke puts forward earnings for the S&P 500 about 20% above the trailing twelve months, a spread that is very strong by the standard of recent history, and the estimates are still climbing. In my September 12 letter I leaned on Bespoke’s Matrix of Economic Indicators, which had swung sharply negative over two months, and I read it as the economy slowing beneath the stock market. I gave that reading more weight than it deserved. The Matrix counts how many year-over-year comparisons are improving or deteriorating. A run of hard comparisons can turn it negative while the economy keeps growing, and I believe that is what happened. The flash US Composite PMI from S&P Global rose to 58.4 in September, a 62-month high and the fourth straight month of accelerating activity. Purchasing managers around the world are reporting much the same. So the economy did not stall in June. Its momentum faltered and recovered, and I mistook the falter for a trend.

An expanding world economy needs higher interest rates, and I no longer read the rise in yields as a threat by itself. The doom scroll reads last week’s hike as the Fed choking the recovery. Bespoke’s work points the other way. The climb in the ten-year has come almost entirely from real yields, while long-term inflation expectations sit close to where they’ve been for years. That is the bond market pricing stronger growth, which is exactly what the bond market should do in an expansion. The inflation that remains is coming from bottlenecks. The same PMI report showed order backlogs building, delivery times stretching, and input costs rising, led by fuel and freight. Diesel is the clearest case, and the diesel crack (the premium of diesel over crude) sits at a record. My view is that these bottlenecks prove temporary. High prices are how an economy tells producers to build capacity, and capacity is being built.

I’m watching two things. The third quarter reporting season begins in the middle of October, and the estimates in the chart above have to be confirmed by what companies actually report and what they guide. And I’m watching the bottlenecks. If input costs are still climbing at year end and inflation expectations break out of the range they’ve held, then the rise in rates is about inflation after all, and I’ll have been wrong to call it noise. Bespoke makes the fair point that with profit margins at records, something has to give. With multiples already back in the high teens, the risk in this market sits in earnings. That is where my attention stays.

Market Technicals

My own technical models are still urging caution, and I intend to respect them. The tape is a different category of evidence from a headline. It’s the market’s record of what investors are actually doing with their money, and I weigh it accordingly. The division of labor is simple. The fundamentals tell me whether to own stocks. The tape tells me when to add, and for three months it has said wait. The Lowry-style distribution I described on September 12 hasn’t been repaired. On Monday the S&P 500 rose better than 1% to close within 1% of its high, yet more of its members made new 52-week lows that day than new highs. One newsletter pointed out that the only two prior examples came in July of 1929 and December of 1999. That is the purest doom scroll material I saw all week, and two observations make an anecdote. Still, the condition it describes is real, and it matches the narrowing in my advance-decline work.

The same week brought evidence on the other side. The Nasdaq closed at a record on Monday for the first time since June 2. Bespoke counts thirteen prior stretches of three to twelve months without a record close, none with a decline of more than 15%, and in the month after each one ended the Nasdaq rose. The semiconductors, the group I’ve called the leadership of this cycle, have turned up sharply this month, as the chart below shows. None of that is a buy signal from my models yet. But it’s the kind of evidence Dan Sullivan taught me to take seriously. Sullivan’s observation was that the first 20% or so of any profitable change in direction happens in the first four or five days. Most investors miss it, because they are waiting for confirmation that arrives after the move has been made.

Magnificent Seven (MAGS), semiconductors (SMH), and data center infrastructure (DTCR), one-year change. Source: Bespoke Investment Group.

So I’m more vigilant than usual. What would change the call is a confirmed buy signal from my own models, with the percentage of S&P 500 members above their 200-day average turning back up through 50% and the cumulative advance-decline line rejoining the index. If those arrive, I’ll act within days. If instead the index keeps making highs on fewer and fewer stocks, the distribution process has further to run and my caution stays in place, however good the earnings look. Long experience has taught me that the tape usually knows something before I do.

Prognostication

My two confirmations fundamental and technical are split again, and this time I know which one governs what. The fundamentals are the signal, and they say the earnings behind the S&P 500 are growing fast enough to carry this market through a Fed that’s raising rates into an expanding world economy. The technicals govern timing, and they say the weeks ahead remain unfriendly. So I hold the direction with conviction and the timing with patience. I still expect the low for this move in the middle of October, the call I’ve made in each of the last three letters, and I expect my models to give a buy signal on or around it. When they do, I won’t spend a week debating it. Sullivan’s four or five days are the reason.

I’ve been thinking this week about Everest. The mountain’s most demanding requirements for success are patience, preparation, and the discipline to concentrate on the goal ahead. Climbers wait weeks at base camp for a weather window, acclimatize in stages, and carry a plan for every hour above the last camp. High on the mountain, distraction can be deadly. Investing asks for the same three things, and the doom scroll works against all of them. It makes patience feel like negligence, it turns preparation into reaction, and it pulls attention away from the goal toward whatever arrived on the phone in the last five minutes. For me the signal is still earnings, and my models are the preparation. So put down the phone, watch the earnings, and keep the buy list current. The climbers who reach the summit are the ones who knew what they were waiting for before they left base camp.

CJ Brott

Chairman Emeritus, Capital Ideas

The material presented is for informational purposes only and is believed to be accurate. Sources include but are not limited to publications by FactSet, Dow Jones, Yardeni Research, and Bespoke Investment Group. All expressions of opinion reflect the judgment of the author as of the date of publication and are subject to change. The author may hold positions in securities mentioned in this commentary.

Zweig’s Second Rule

September 19, 2026

Marty Zweig left investors two rules and most people only remember the first one. Don’t fight the Fed, they forget don’t fight the tape. He wanted both satisfied before he would own stocks with conviction. That is the discipline I describe here as my two confirmations fundamental and technical. Kevin Warsh raised rates a quarter point on Wednesday, the first hike in three years. Followed blindly, the first rule now reads plainly against equities. The question this letter asks is whether I should obey it. Zweig’s rule was never about the funds rate by itself; it was about the economic structure that carries the cost of money into corporate earnings and household spending. And the rule only works as well as that system holds. My view is that the mechanism is no longer as workable for the current market as represented by the S&P 500 index. So, the second rule, the one about the tape, is the one governing my behavior this week. The near-term history is against me, and the chart below shows it clearly.

Average S&P 500 and sector performance one month after the first hike of a tightening cycle, six cycles since 1994. Source: Bespoke Investment Group.

Market Fundamentals

The Fed tightened into an economy that is still growing, and that part of Zweig’s first rule I take seriously. The committee voted unanimously on Wednesday and futures market pricing points to at least one more increase before year end. That would make this a cycle rather than a gesture. Bespoke counts it as the seventh tightening start since the modern Fed began in 1994, and in the month after the first hike of the prior six the S&P 500 fell on all but one occasion. Financials took the worst of it and technology barely moved. Six observations is a small sample and I wouldn’t build a portfolio solely on it. But the pattern is consistent enough that ignoring it would be careless. I’m cautious over the next several weeks and the Fed is a good part of the reason.

Where I part company with the rule is on what a hike can actually reach. Rate increases work on the economy by raising the cost of borrowing and by squeezing consumer households. The S&P 500 we own today is not built out of borrowers and households the way it was when Zweig was writing. Bespoke puts the combined weight of the two consumer sectors in the S&P 500 at 13.5%, down from better than 30% before the dot-com boom. The lowest reading on record. The companies that have taken their place fund capital spending out of operating cash flow and sit on net cash. That makes a quarter point on the short rate close to irrelevant for their earnings power. Index earnings are growing at roughly 28%, driven by that same group, and I can’t see the mechanism by which Wednesday’s move slows that materially over the next two quarters. The consumer, meanwhile, is still spending, with August retail sales well ahead of forecast. He may slow from here, and if he does the damage lands on the economy well before it lands on the index.

Inflation is the piece of this that won’t resolve on the Administration or the Fed’s schedule. It runs directly through oil. Crude finished the week above one hundred dollars, and the more damaging number is in the products, where diesel trades at an extraordinary premium to crude. Renewed strikes on Russian refining, and refinery outages here at home, contribute to the problem. Almost everything an American consumer buys spends part of its life on a truck, so a diesel price at these levels is a tax that directly affects the price of goods. Of course, that is on a lag and holds the year-over-year comparisons up. That argues for higher rates for longer no matter what the monthly data prints. Adding to that, the bond market agrees, with the two-year to ten-year spread flattened to the low end of its range for the year. I’m watching the diesel crack and front-month crude more closely than anything out of Washington. If crude comes back under ninety and those premiums narrow, the inflation problem eases. The October meeting becomes a genuine question rather than a foregone conclusion, and the fundamental case improves quickly. If instead, the large technology companies guide their capital plans lower in October because money has gotten expensive, then the tightening Fed policy effect stays intact, and both maxims are clearly in agreement.

Market Technicals

The tape is the rule that has been in effect, and it hasn’t improved. Last week I wrote that participation was draining out of a market whose index level had barely moved, and another week of trading added to that evidence rather than subtracting from it. The Dow Transports broke below their 200-day moving average on Friday for the first time in nearly a year. The semiconductors, which Bespoke calls the “transports of the 21st century” and which I treat as the leadership group of this cycle, spent the week pressed against their 50-day average and after multiple attempts since July failed to clear it once more. The major banks and brokers all finished below their 50-day averages after a hard week, which is the sector the first-hike record says gets hit hardest, and they weakened as expected. The equal weighted S&P 500 index has broken the rising channel it held from April through August. There’s one exception to all of this, and it is the reason I currently expect to be a buyer of stocks later this year.

The exception is the group the Fed cannot reach. The Mag 7 ETF made a new all-time intraday high on Friday while the large cap ex-Mag 7 ETF closed solidly below its 50-day average for the first time since the Iran war began. I read that divergence as the tape agreeing with the fundamentals. Money isn’t leaving this market so much as concentrating in the companies whose earnings don’t depend on the cost of money, which is what you’d expect to see if Zweig’s first rule has lost its grip on part of the index. It’s the least reassuring way to remain bullish, because narrow leadership is still narrow leadership. Still, I’m a strong believer in letting the market tell me what to do rather than me telling the market what to do.

Mag 7 (MAGS) versus large cap ex-Mag 7 (XMAG), total change since October 2024. Source: Bespoke Investment Group.

The key is the semiconductor index and its 50-day average. If the semis clear that line and hold it, the rest of the tape has a path higher and I’ll read this month as a “correction in time” rather than the front edge of something worse. If instead the Mag 7 roll over and join the transports and the banks below their own averages, the one group standing between this market and a real decline has given way. And I’ll act accordingly as long experience has taught me that market prices move well in advance of economic fundamentals.

Prognostication

My two confirmations agree about the next month and disagree about later in the year. The Fed and the tape both say the weeks ahead are unfriendly, which is why I still expect the low for this move in the middle of October, the same call I made in the last two letters. The two, technicals and fundamentals, differ on what that low will represent. The tape, read strictly, describes a distribution process that could carry further than an ordinary seasonal decline. The earnings data describes a pause inside an expansion that’s still intact. When my confirmations split this way the honest response is to hold the timing with conviction and the severity loosely, which means I’m a buyer into October weakness and never a seller into it.

Sentiment supports that posture. AAII bearish readings crossed back above 50% this week to their highest level since the tariff tantrum of 2025. The bulls fell under 30% for the first time in a year. Readings at those extremes have been a reliable intermediate-term contrarian signal for most of my career.

AAII bearish sentiment, weekly, 2021 to 2026. Source: Bespoke Investment Group.

The year to look at is 1994. Greenspan raised rates for the first time in five years on February 4 of that year and then doubled the funds rate over the following twelve months, into an accelerating economy. The bond market absorbed nearly all of the damage, Orange County and the Mexican peso broke, and equity investors got twelve months of chop that ended near where it started. No economic contraction, no bear market, just a year of frustration, and then 1995 returned better than 34%. That’s the shape I expect here, a sloppy market working out whether the Fed has made a mistake, resolving higher once earnings season shows the companies carrying the index still compounding. Coincidentally, 1994 was a midterm year as well.

My list of demands going into October is short: crude back under ninety with a narrower diesel crack, the ten-year below five percent, the semis clearing their 50-day average and holding it, and the third quarter capital spending guidance that starts arriving in the middle of the month. I’d add AAII bulls back above 40%, which would tell me the contrarian case has been spent. Until they show up I intend to do very little, keep my buy list current, and buy weakness rather than wait for an all clear that never sounds until prices have recovered. Stay invested, stay diversified, and don’t fight the tape. Zweig’s first rule will have my attention again when the Fed can reach these earnings, and not a day before.

CJ Brott

Chairman Emeritus, Capital Ideas

The material presented is for informational purposes only and is believed to be accurate. Sources include but are not limited to publications by FactSet, Dow Jones, Yardeni Research, and Bespoke Investment Group. All expressions of opinion reflect the judgment of the author as of the date of publication and are subject to change. The author may hold positions in securities mentioned in this commentary.

September 12, 2026

The men who flew airplanes into the World Trade Center twenty-five years ago yesterday morning were not attacking office space. They were attacking a system, and they said so plainly. Osama bin Laden spoke repeatedly of bleeding the American economy, and the towers were chosen for what their name announced: world trade. I have thought about that a good deal this year, because 2026 is also the year the country turns 250, and the two anniversaries sit strangely together, one a day of funerals and the other a year of parades. What strikes me is that they point at the same object. In 1776 two documents appeared four months apart. In March, Adam Smith published The Wealth of Nations, which described how self-interest coordinated through voluntary exchange produces a prosperity no planner could design. In July, Jefferson’s Declaration asserted the personal liberty that makes the pursuit of that self-interest a right rather than a privilege. One supplied the machinery and the other supplied the permission. Bin Laden aimed at what those two documents built.

He failed, and the ledger says by how much. Bespoke notes this week that the total return of the S&P 500 from the close on September 10, 2001, the last session before the attacks, through Thursday is 1,007%, an annualized gain of 10.1%. Better than a ten-bagger. It was not a straight line and I want to be clear about that, because the full story is the more useful one. Equities fell better than 11% in the first week back, the ultimate low of the dot com bust did not arrive for another thirteen months, and an investor who bought the reopening sat through a further decline of roughly a quarter before any of the compounding began. The lesson is not that markets shrug off catastrophe. It is that a system resting on those two documents is distributed across three hundred million people making their own decisions, which is why the exchanges reopened in six days with lower Manhattan still burning, and why the compounding resumed. That is the long view, and I hold it as firmly as I hold anything. What follows is the short view, and the short view this week calls for care.

Percentage of S&P 500 members above their 50-day and 200-day averages. Source: Bespoke.

Market Technicals

Bull markets do not end on a day. They end through a process, and I believe that process has been underway since June. I have leaned on the Lowry system for most of my career, and its central insight, published continuously since 1938, is that major tops are built by a gradual withering of demand that shows up in advance-decline statistics long before it shows up in the averages. Distribution is a process, not an event. Earlier this year better than 70% of S&P 500 members traded above both their 50-day and their 200-day moving averages. As of Thursday, per Bespoke, 55.5% remain above the 200-day and only 37.7% above the 50-day, the worst readings since the spring. That is not a wobble. That is participation draining out of a market whose index level has barely moved.

The sequence matters more than the levels. The leadership complex topped first: memory stocks peaked on June 22 and fell more than 40% into late July, the semiconductor index has traded under a downtrend line from its June high ever since, and the Nasdaq never made a new high in August at all. The S&P 500 did make a marginal new high on August 13, but it made it on the shoulders of a handful of mega caps while the equal weighted index and the Russell 2000 broke their 50-day averages and kept going. The cumulative advance-decline line, which I treat as a leading indicator rather than a confirming one, has now broken its uptrend and sits more than two standard deviations below its 50-day average, a first percentile reading going back to 1990. That is precisely the divergence Lowry teaches you to respect, the index holding its ground while the market beneath it narrows. I am watching the percentage above the 200-day above all else. If it holds over 50% and turns back up while the index consolidates, I will read the last month as rotation rather than distribution, and I will be the first to say so. If it breaks 40%, the process is further along than I currently think.

Market Fundamentals

The market did not slow on its own. It slowed because the economy did, and the timing lines up too neatly to set aside. Bespoke’s Matrix of Economic Indicators counts how many measures are accelerating year over year, and through May the reading had climbed to plus 21, the strongest since 2024 and the fourth consecutive positive month. By July it was minus 13. That thirty-four-point reversal over two months is the third largest since 1999, behind only March 2020 and June 2021, and it began in June, the same month the leadership groups rolled over. Housing carried much of the damage, with starts down better than 13% from a year ago, and six of seven consumer measures deteriorated. Manufacturing was the lone bright spot. This is a deceleration in momentum rather than a stall, and I want to be careful not to overstate it, but a market narrowing at the same moment the data turns gives me two separate arguments arriving at the same conclusion.

What makes it harder is that the two things which actually price equities have both moved against us. Crude crossed $100 this week. The ten year Treasury sits within a hair of its fifty two week high, and two-year yields have added nearly forty basis points since Chairman Warsh spoke at Jackson Hole on August 28. The market now prices a rate hike at next week’s meeting at 87%, up from 60% on Wednesday. High oil and high yields are simply a headwind to equity prices, and until at least one of them relents it will be difficult for this market to do much more than churn. A central bank tightening into decelerating data is the 1994 problem I wrote about two weeks ago, and it remains the single largest risk I see between here and year end. If crude rolls back under ninety and the ten-year backs away from its highs, the fundamental case improves in a hurry and my technical caution will have amounted to a month of chop. If both keep climbing while the Matrix stays negative, both arguments point the same way and I will position accordingly.

Prognostication

I expect the low for this move to arrive in the middle of October, and I expect to be buying into it rather than selling out of it. Two weeks ago I wrote that fundamentals and technicals were handing me a split decision, and that the right answer was Richard Russell’s: don’t just do something, stand there. They are no longer split. Both now point the same way for the near term, and when my two confirmations agree I take them seriously. That does not make me bearish. It makes me patient, and patience in September has a specific shape to it. Bespoke’s work shows that virtually all of the month’s historical weakness comes in the second half, and that the average daily move keeps expanding until roughly October 20 before settling down into year end. We are entering the part of the calendar that does the damage, not leaving it.

I have watched markets long enough to trust a rhythm that most people dismiss as folklore. Markets almost always seem to bottom in mid-October. The crash low came on October 19 of 1987, the Gulf War low on October 11 of 1990, the Long Term Capital low on October 8 of 1998, the dot com low on October 9 of 2002, the European crisis low in the first days of October 2011, and the most recent bear market low on October 12 of 2022. Six of them, across six entirely different causes. I do not know why it clusters there and I would not bet a portfolio on a date, but I have seen it too many times to plan around anything else. What I intend to do is let the deterioration run its course, keep my list ready, and put money to work into weakness rather than waiting for the all clear, which in my experience never sounds until prices have already recovered.

The other reason for patience is what sits on the far side of October. This is a midterm election year, and in midterm years the twelve months following the September 30 close have averaged a gain of 19.47% since 1945, better than twice the average for all years. I would not lean on a seasonal statistic by itself, but it points the same direction as everything I believe about the decade, which is that the capital being poured into artificial intelligence today is the installation phase of a general purpose technology whose payoff arrives on a lag we cannot schedule. If mid-October passes and the percentage of S&P 500 members above their 200-day average is still falling, I will have been wrong about the timing and I will tell you so. Twenty five years ago the exchanges reopened in six days and went on to return better than ten times over. Two hundred fifty years ago two documents set that machinery in motion. Neither anniversary tells me a thing about October. Both tell me exactly what to do about the decade.

CJ Brott

Chairman Emeritus, Capital Ideas

Chart from Bespoke Investment Group. Other data from FactSet, Dow Jones, Yardeni Research, and Bespoke. This letter reflects my personal opinions as of the date above and is for informational purposes only. It is not a recommendation to buy or sell any security and is not personalized investment advice. Past performance is not a guarantee of future results.

Don’t Just Do Something, Stand There

September 5, 2026

The chart below is the reason I am not selling into this dull, narrowing market. Bespoke has tracked the Nasdaq since the release of ChatGPT against the Nasdaq after the release of the Netscape browser in 1994, and the two lines have stayed remarkably close for almost four years. By their count we are now at September 14, 1998. That was the fall of the Asian crisis and the collapse of Long-Term Capital Management, a frightening leverage unwind that took the Nasdaq down hard into October. The eighteen months that followed were the strongest of the entire decade. The question this letter tries to answer is why the market should repeat that pattern rather than break from it. My answer comes in two parts. Technically, what I see is a “correction in time” inside a trend that is still intact. Fundamentally, the growth ahead from the adoption of artificial intelligence is larger than the market is giving it credit for, and it has barely started. Long time readers know I am most confident when those two sets of tools agree. This week they mostly do on the long view and do not yet on the short one, which is why my advice is patience.

Source: Bespoke Investment Group, The Bespoke Report, September 4, 2026.

Market Technicals

The 1998 correction was caused by leverage, not by any failure of the technology, and that is the first thing the present tape has in common with it. Leverage always leverage, as Scott Nations’ five crashes remind us. This cycle’s leverage flush so far was Situational Awareness in August, and the Nasdaq has been calmer since it left the market. If a bigger one is coming it will show up first in credit spreads and the St. Louis Fed’s stress index, and neither is saying anything alarming. The 200-day moving average is still rising smoothly, so on trend I remain a bull. What I see underneath is the market resting. The S&P 500 is within half a percent of its high, but the percentage of stocks above their 50-day moving average fell below 50% on Tuesday for the first time since May, and the cumulative advance-decline line has more ground to make up than price does. Bespoke thinks this divergence now only reflects the strength of the biggest companies. I am less relaxed about it, because the weakness sits in the leaders. The semiconductors, which I have long called “the transports of the 21st century,” and the actual Dow Transports are both in downtrends of lower highs dating to June.

The rotation out of the industrials is the clearest example of a correction in time, and also of how confusing this market is to read. Industrials made new highs in mid August and then broke below the 50-day without stopping until they neared the 200-day. Bespoke sorted the sector by each stock’s correlation to Nvidia over the past two years. The industrials least tied to Nvidia barely fell. The ones that trade with it, the electrical equipment, cooling, and power companies building the data centers, fell about 12%. So the strongest part of the economy is the part of the sector being sold hardest. That is not a verdict on the buildout. It is a crowded trade being trimmed while investors wait to see whether the companies buying all that equipment will earn a return on it. The same thing happened to the technology sector for eighteen months in 2020 and 2021, and it resolved with new highs. Bespoke’s “three-headed monster” of the dollar, oil, and the 10-year yield explains why the rest has to wait. The dollar has gone back into its cave. Oil and rates have not, and the market will not make its next leg higher until one of them joins the dollar. Sentiment is split the same way: individual investors are scared with AAII bears well above average while price sits near a record, which by my reading is fuel, and the professionals look complacent. This is the “exhaustion” I described last November, when neither side could hold a conviction until a catalyst arrived. I am watching the equal weight S&P 500, which ended the week on its 50-day at the bottom of its uptrend channel. If it breaks that channel while the advance-decline line makes a lower low, that is a correction in price, not time, and I will say so. If a mild inflation report brings breadth back above 50% and the semiconductors break their downtrend, the tape confirms the long view and I will get more aggressive.

Market Fundamentals

The Fed is the near term risk to the pattern and the economy is the long term reason it holds. Last week I wrote about the question Chairman Warsh left open at Jackson Hole. This week Governor Waller answered half of it: he will vote to hold if the August inflation report on September 11 shows continued progress, and consider a hike if it comes in hot. Then Friday’s payroll report came in at about three times what economists expected and the odds of a hike went back to a coin flip. Marty Zweig’s first rule is don’t fight the Fed, and I have followed it for fifty years. But the Fed has not moved. It has held rates at a level Waller calls only slightly restrictive for five straight meetings, and the year I keep coming back to is 1994, when Greenspan began raising rates into an accelerating economy, the market went nowhere for a year, and 1995 became one of the best years in history because the strength that caused the hikes was real. A hike into strength is a delay. A hike into weakness is a danger. If inflation comes in hot, the Fed hikes, and the 10-year Treasury moves through 5%, the pattern gets a real test. If inflation cools and Waller’s hold carries, the near term risk goes away.

The economy underneath the Fed debate has one very strong engine and several idling ones. Dell, Hewlett Packard Enterprise, and Broadcom all reported this week and each said the same thing: demand for AI servers and chips is running well ahead of plan. Data center construction has passed the peak office construction reached in 2020 while nearly every other kind of building has been sliding since 2024. Bespoke’s AI baskets show the same split. The infrastructure basket, the companies selling the equipment, is up over 60% this year, and the implementation basket, the companies that are supposed to use AI to make money, is roughly flat. The Wells Fargo chart below shows why that gap matters. AI capital spending as a share of the economy is already the largest since the railroads of 1865 to 1890, and the railroad boom produced the fastest fifty years of income growth in American history because it lowered the cost of moving goods and moving information at the same time. That payoff came from adoption, from what the country did with the railroads once they were built. The growth ahead from AI is the same kind, and it only arrives with implementation. What no one can tell you is the lag. The electric motor was in factories in the 1880s and did not change productivity until the assembly line in the 1910s. The market is pricing the buildout as if the lag were knowable, and it is not, which is exactly why the implementation names are flat and the equipment names are being trimmed. I am watching for the first quarter in which the implementation basket outperforms the infrastructure basket on earnings rather than on rotation. That will tell me the payoff has begun.

Source: Wells Fargo Investment Institute, “Can AI help pay America’s bills?”, September 3, 2026.

Prognostication

This week I sat with a client whose money I have managed for several decades. He had been shown an investment proposal by another firm. It was complicated, it called for major changes, and it was not suited to his temperament. He turned it down, and I told him he was right to. In the 1970s Richard Russell, who wrote the Dow Theory Letters for more than fifty years, used to say “don’t just do something, stand there.” He meant that most of the damage investors do to themselves comes from acting during dull stretches because doing nothing feels like neglect. I have been in this business since 1970, and a flat market with the Fed undecided and breadth narrowing is exactly the stretch Russell was talking about. The temptation to rearrange a portfolio is strongest when nothing is happening. The right response is to know what you own, know why you own it, and wait for the market to tell you something.

The calendar agrees. September is the only month that has averaged a loss of more than 1% since 1928, but in the 36 years when the S&P 500 was already up more than 10% through August, the last four months averaged a gain of 4.7% with gains in 30 of them, and midterm years add a second tailwind. My base case is that the standoff lasts through the Fed meeting and resolves during October earnings, when the companies building the second wave report again. I expect the S&P 500 to finish 2026 higher than it is today, with the equal weight index and value stocks leading and the semiconductors needing a rest before they lead again. The valuation objection is real, with the S&P dividend yield below 1% for the first time since 2001 while the 10-year pays nearly 5%. The 1990s answer to that objection was earnings growth, and it is the same answer now. For early 2027 the question becomes whether the implementation companies begin to show the payoff, and that is the year I expect the first real evidence. What I am watching next: the inflation report on September 11, the Fed decision on September 16, the equal weight channel, and whether the semiconductors and the transports break their downtrends. What would change my mind is all three together: a rate hike, a 10-year yield through 5%, and a breakdown in breadth. Any one alone is noise. Until then, stay invested, stay diversified, and don’t just do something, stand there.

CJ Brott

Chairman Emeritus, Capital Ideas

The material presented is for informational purposes only and is believed to be accurate. Sources include but are not limited to publications by FactSet, Dow Jones, Yardeni Research, and Bespoke Investment Group. All expressions of opinion reflect the judgment of the author as of the date of publication and are subject to change. The author may hold securities mentioned in this commentary. This memo does not constitute investment advice, and readers should consult with their financial advisors before making investment decisions.

Start with where the new Chairman started. In his first Jackson Hole address, Kevin Warsh opened not on inflation, not on the labor market, and not on his own principles for the job, but on artificial intelligence. He called this a hinge point in history and described the technology as potentially a new factor of production, which is uncommonly strong language from a sitting Chairman. Then he did two things that belong side by side. He handed the questions that actually matter, whether this produces a sustained rise in productivity and when, whether it complements labor or competes with it, and where the returns finally land, to a task force, and said plainly that its findings will have no bearing on current policy. And he declined to say anything at all about where interest rates are headed. A Chairman who will not forecast rates spent the opening of the most watched speech he will give this year on the technology. He did not endorse the payoff and I will not pretend he did. But the placement was the message, and the questions he left open are the ones this letter has been answering for two years.

He did not need to forecast, because the rest of the world is forecasting for him. Long term yields are rising in the United States, Germany, France, the United Kingdom, Canada, and Australia at the same time. Those six countries have almost nothing in common that bears on this. Their fiscal positions run from disciplined to alarming, their short rates are set by four different central banks, and their politics do not resemble one another. The explanation that fits all six at once is the plain one: more of the world wants to borrow. This is not an inflation scare, since long run inflation expectations have been quiet for years, and it is not a flight from bonds, since the extra compensation investors demand for holding long paper has barely budged. It is the price of capital rising because the demand for capital is rising, which is the precise opposite of the savings glut that every serious economist described after 2008. For fifteen years the complaint was that there was nothing worth funding. Something is now being funded, everywhere at once.

Ten year government bond yields, latest versus twelve months earlier. Source: Trading Economics, August 28, 2026.

This is the fact the bear case cannot absorb. Michael Burry and the Man Group have moved past arguing that the demand for artificial intelligence is fake. Their case now rests on financing mechanics: depreciation schedules that assume a five or six year life for hardware that may last two or three, circular commitments in which one company’s revenue is another company’s capital pledge, lease obligations parked off the balance sheet, and free cash flow compressing across the largest technology companies. I take those seriously and I do not dismiss them. But they are arguments about how the buildout is being paid for, and arguments about financing are settled in the end by growth. Five firms can trade paper with one another. They cannot lift the cost of long term money in six sovereign bond markets at once, and they certainly cannot do it in Ottawa and Canberra.

Al Root made the scale argument in Barron’s this week and it deserves more attention than it got. Looking across 250 years of American capital booms, he finds the economy has absorbed spending on a transformational technology equal to roughly a quarter of a year’s national output before things turn ugly. Railroads were the template: annual output near ten billion dollars in the 1860s against rail spending that eventually totaled two and a half billion before the panic of 1873 arrived. Apply that same yardstick today, with output near thirty trillion, and the danger zone sits somewhere around seven and a half trillion dollars, which is another five to six trillion of domestic spending from here. At the current pace that line does not get crossed until the early 2030s, six or seven years into this boom. I would not lean on the decimal places, since that is Barron’s own construct rather than a standard series, but the order of magnitude is what matters. There is room, and room is time. Time is precisely what converts a buildout financed by other people’s money into one that pays for itself.

What tells me the growth is real is that it does not look like a software story. Japanese machine orders for industrial robots are up 26.4 percent over the past year, the fastest in five years, with machine tool orders accelerating on demand from China and the emerging markets. Korean manufacturing sentiment has reached new cycle highs on memory demand and is spilling into the rest of that economy. Germany’s benchmark index and Sweden’s both closed at records on Friday. Factories in countries that own none of the underlying intellectual property are running harder because of it. That is what a general purpose technology looks like while it is being installed, and it is what electrification looked like between the wars. The dynamos went in during the 1890s and 1900s, and the productivity gains did not arrive until the 1920s, after plants were rebuilt around unit drive instead of a central shaft. I made that argument in my January 2021 letter under the heading of a “Roaring 2020s” and I make it at length in the book this fall. Installation first, integration a decade later. I am watching for the one thing that would change this reading, which is capital spending growth decelerating while the economies around it stall. A worldwide industrial expansion and a technology this consequential arriving together may simply be more than the accountants can stop.

Market Technicals

The tape agrees with the thesis and disagrees with its leadership, and that tension is the most useful thing on my screen. Since earnings season began on July 13, Microsoft and NVIDIA together added roughly 1.41 trillion dollars of market value while the other seventy one Technology stocks in the index lost 22.3 billion dollars combined. Two companies accounted for the entire sector gain. Concentration of that order is a condition rather than a sell signal, but it does mean the index is currently reporting on two balance sheets rather than on an economy. It also sits oddly beside a genuinely broad summer: from May 18 through August 21 the equal weighted S&P gained 9.3 percent against 3.7 percent for the cap weighted index, and then the market re-narrowed violently in the last two weeks of earnings season. Both things are true and I see no reason to resolve the tension. Breadth improved while the leaders rested, and the leaders took it back when they reported.

Beneath the leadership the market is more discriminating than the headlines suggest. Bespoke’s basket of picks and shovels names, the contractors, power providers, and turbine makers levered to the data center buildout, has been in a steady downtrend since April. At the same time, the basket of companies most exposed to being disrupted by this technology, the ones that collapsed in the first quarter on fears that credit reporting and food delivery and office leasing were about to be made obsolete, has recovered every bit of that loss and now trades at its best level since the tariff shock of 2025. Read those together and the market is making a specific judgment: the infrastructure is real, the displacement story was oversold, and the enthusiasm for anything with a data center attached is being rationed.

The week also handed me my standing thesis with a name attached to it. Situational Awareness, a hedge fund carrying what Bespoke fairly called absurd leverage in high beta artificial intelligence names, had to sell its entire public equity book to Citadel in August after that group broke. “Leverage always leverage” is the phrase I keep returning to, and Scott Nations is right that the great breaks in our history trace to new vehicles combined with borrowed money rather than to the underlying idea being wrong. A leveraged fund blowing up inside a theme is not the theme failing. It is the tuition. I am watching the percentage of the S&P 500 above its 200-day moving average, which stood at 72 percent in my August 15 letter with a thrust through 75 percent as the signal I want, along with the cumulative advance-decline line and whether the 200-day is still rising smoothly. What would change my reading is credit. If spreads on the data center and neocloud financing complex widen while investment grade stays calm, that marks this buildout as the source of the stress rather than a passenger in a general risk-off, and I would act on it.

Prognostication

The cost of capital is rising against a capital intensive thesis, and I do not think that ends the thesis. The market moved the odds of a September hike to 57.1 percent on Friday from 35.8 percent the day before, and futures now carry roughly two increases by the spring of 2027 with declines after that. Warsh would not forecast and the market forecast for him. That is a policy rate catching up to a neutral rate the world’s demand for capital has already pushed higher, not a campaign to break the back of demand the way the early 1980s were. Raymond James strategist Tavis McCourt, quoted in the same Barron’s piece, makes the point that these booms rarely die of their own accord and usually need an outside shock to end them, often a central bank. That is the risk worth naming, and I name it. A Federal Reserve that tightens into a capital cycle it does not yet understand is the most plausible way this gets interrupted.

Interrupted is not the same as ended. What has to happen for a capital spending boom to succeed, in McCourt’s own framing, is that the outlays slow while profitability ramps, so the businesses become self-financing. There are early signs of exactly that. Root notes that a gigawatt of computing power costing forty billion dollars to build can generate something close to forty billion in annual sales, against an S&P 500 that needs two to three dollars of assets to produce a dollar of revenue. If those economics hold even approximately, the cash flow arrives before the financing runs out, and the profits follow the cash flow. McCourt is honest that a smooth ramp to profitability without hiccups has never happened at this scale outside of government funded booms, and I will not wave that away. But he is describing a race between growth and financing, which is the right way to frame it, and everything in the global data this week says growth is running well.

So I hold two things at once, which is unfashionable but honest. Near term I expect turbulence, and I would not be surprised or alarmed by a meaningful drawdown given how much of the index now rests on two companies and how much borrowed money is chasing the same theme. Longer term I think we are early in the installation phase of the most important capital cycle since electrification, and a worldwide industrial expansion is supplying the demand that carries it. I will change that view if the hyperscalers shorten the stated useful lives on their hardware, which would validate the depreciation case and reset earnings across the group, if the picks and shovels downtrend spreads to the buyers rather than staying with the sellers, or if capital spending growth slows while the profitability that is supposed to replace it fails to appear. Absent those, the Federal Reserve can raise the price of money. It cannot repeal a capital cycle.

CJ Brott, Chairman Emeritus, Capital Ideas

Data and charts referenced in this letter are drawn from FactSet, Dow Jones, Yardeni Research, Bespoke Investment Group, and Trading Economics. This letter is for informational purposes only and does not constitute investment advice or a recommendation to buy or sell any security. Past performance is not a guarantee of future results.

August 21, 2026

This is the first of these letters in a new place, so a word about what it is.

I have spent more than five decades in the securities business. I founded a brokerage firm and a registered investment advisory firm, and for thirty-five of those years I also taught investing and financial planning at Southern Methodist University, running the firms during the week and teaching on the side. What that combination taught me is that the point is not to predict the market. It is to recognize the patterns that repeat as one technology gives way to the next.

Rather than assert that, here is a passage from a client letter I wrote in January 2021.

That may all sound like a “this time it’s different” justification for a continuation of this secular bull market. This time probably is different. The paradigm for that logic is the current new “roaring twenties” model. Over the last ten years we have discussed this idea with clients and alluded to it in our letters. Our logic depends on an understanding of the post-Civil War economic history of the United States. During the period from 1870 thru 1895 many inventions were patented ranging from crop harvesters to railroad air brakes. The telephone was invented and practical methods for generating and distributing electricity were developed. However vast sums of capital were misallocated and lost on technologies that had no mass market applications. It was not until the 1920’s that telephones, electric lights, refrigeration and mass-produced automobiles were assimilated into the economy and economic changes truly made it “different this time.”

Earlier in that same letter I named what I thought was being absorbed: the internet, supercomputing, genetic engineering, and the application of artificial intelligence computing across all fields and industries. That was January of 2021, before most people had heard of a large language model.

I do not include it to take a victory lap. The letter got plenty wrong, and I will say so when it comes up. I include it because it shows what this letter is. Not a forecast of next quarter, but an argument about where we sit in a cycle that runs for decades.

The thesis has not changed. Invention comes first, capital gets misallocated for years chasing applications with no market, and the returns arrive only when the technology stops being a product and becomes infrastructure. That is what happened with electricity between 1915 and 1940, and it is what I believe is happening now with artificial intelligence. The longer version became a book, Investing in the Second Wave, published this fall by Greenleaf Book Group.

Which brings us to this week, where you can watch the assimilation happening in real time.

The Economy

The economy is running hotter than most people realize. The Philadelphia Fed survey, taken every month since 1968, just delivered one of the seven best readings in its history, with the six month outlook the strongest since 1983 and capital spending plans near record territory. You do not get numbers like that unless businesses believe the orders are coming, and this is the manufacturing sector we spent the better part of a decade writing off.

Keep one wrinkle in your back pocket. Good news for the economy and good news for stock prices are not the same thing. When that survey has run this hot before, the market has generally struggled over the following year. Peak confidence in the factories tends to arrive alongside peak pricing power, and multiples do not expand into that.

Housing is the one clear weak spot, and the trouble is supply rather than demand. Existing owners are welded to mortgages they will never see again, so nothing comes to market.

On rates, long yields are rising on real yields, not on fear of inflation. The five year forward five year breakeven sits around 2.25%, right where it has been since 2021. That is a market voting for a healthier economy and a credible Fed, which is a very different thing from a market losing faith.

Market Technical Conditions

The S&P closed Friday at 7,674 with the VIX at 15, and on the surface nothing happened. Underneath, plenty did. The cap weighted index made a new high, backed off, and found support at the top of its old range. The equal weighted index made a high of its own and faded. The Nasdaq confirmed neither, held back by technology, which has become everybody’s source of funds. Money is being raised in tech and spent in health care, energy, and industrials.

That is rotation, and it is the most important signal on the board. Semiconductors and memory sold off, bounced, and stalled at their declining 50-day averages. While that happened, health care traded to record highs and biotech broke out of a base it had been building for years. Leadership changed hands without the index going anywhere. That is not distribution. A bull market that can hand the baton from semiconductors to health care without losing a step has something left.

Industrials are the piece I would not overlook. Strip the AI story down to what the money is actually buying and you find transformers, switchgear, turbines, cooling systems, and grid. The headlines say technology but the invoices read heavy industry, and the buildout is measured in trillions over the next five years. If you want evidence that we are in the assimilation phase rather than the invention phase, that is it.

My two primary signals both confirm. The cumulative advance decline line made new highs again last week, and it has rarely let a major top happen without warning first. The 200-day moving average is rising smoothly, which after fifty-six years remains the most reliable trend filter I know.

Which brings me to the title. The dispersion index measures how independently individual stocks are expected to trade, and it hit one of the highest readings on record last month. It has since collapsed, one of the sharpest four week declines in its history, and the handful of prior instances all resolved higher, and materially so. Falling dispersion from a record high is the market’s way of saying participation is about to broaden.

Add it up and we have a correction in time rather than a correction in price. The index goes sideways, leadership rotates, and valuations get worked off by earnings rather than declines. A slower and far more pleasant way to reset a market, and the kind of tape that rewards patience and punishes the impulse to do something.

Prognostication for the Rest of 2026 and Early 2027

Mark Twain ranked September high on his list of peculiarly dangerous months in which to speculate in stocks, and the arithmetic backs him up. Going back to 1945, the stretch from here to year end has typically included a drawdown of around 6% somewhere along the way. That is the toll for the trip. But it has been more common for the market to sail through this window than to break in it, and the median finish has been a gain.

The pattern work agrees. Since 1928, the years whose trading most resembles 2026 consolidated in late summer and then finished strong, gaining ground about four times in five. Those years were also up more than twice as much as we are by this point on the calendar. We have been modest so far, and modest leaves room.

So: a choppy September, a firmer fourth quarter, and a market entering 2027 considerably broader than it entered 2026. Earnings, not sentiment, will do the work.

The risk, as always, is leverage. It is always leverage. Every serious break I have watched came from a new instrument combined with borrowed money, and the current candidate is sitting in plain sight over in the crypto complex, which just had its best week in two years. Not a forecast, just a place to keep one eye while the rest of the market gets healthier.

That is the format, and it will not change much: the economy, the market’s technical condition, and where I think this goes. Written the way I taught and the way I talked to clients, directly and without the hedged language that lets a forecaster be wrong in both directions at once.

Replies come straight to me and I read them.

CJ Brott
Chairman Emeritus
Capital Ideas

The material presented is for informational purposes only and is believed to be accurate. Sources include but are not limited to publications by FactSet, Dow Jones, Yardeni Research, and Bespoke Investment Group. All expressions of opinion reflect the judgment of the author as of the date of publication and are subject to change. The author may hold securities mentioned in this commentary. Nothing in this report should be construed as investment advice and does not take into consideration your specific situation. All investments involve risk. Past performance does not guarantee future results.