Markets give you a lot to react to and not much context. This monthly note is our attempt to provide some of that context. Thank you for reading.
If you’re a client, we sincerely appreciate your business. If not, and you would like to learn more about our firm, please feel free to reach out. You’re also welcome to share this update with others who may find it useful.
August 2026 in Review
August saw risk assets, including most stock indexes, resume their march upward and onward. Meanwhile, the repricing of Federal Reserve expectations pulled the rate-sensitive corners down, including small-cap stocks, real estate, and some bonds.
Commodities (+6.2%) tracked the war in Iran. Oil held in the mid-$80s to low-$90s all month, and the national average for retail gas was above $4 every day, setting a new August record. Gold (+9.9%) had its strongest month of the year, though it remains up only 3.1% for 2026 after peaking in January. Speculative positioning in gold futures rose for a fourth consecutive week to its highest level since December 2024. Bitcoin (+25.2%) was the month’s biggest gainer but is still down 10.7% on the year.
In equities, the Nasdaq 100 (+4.2%) rebounded from a weak July. Software outpaced semiconductors for a second straight month. International developed markets (+3.3%) and emerging markets (+3.0%) both beat the S&P 500 (+2.7%). The S&P set record highs mid-month before easing back in the final two weeks. Sector participation was wide with energy, materials, and healthcare all contributing alongside technology. Utilities and industrials were the detractors due to interest rates and AI capex concerns.
The three weakest asset classes tracked were REITs (-2.5%), small caps (-0.6%), and municipals (-0.2%). Core bonds (+0.4%) managed a small gain even though the 10-year Treasury finished the month at 4.75%. High yield bonds (+0.9%) beat investment grade corporates (+0.4%).
The year-to-date column is worth a look relative to the one month. Core bonds (-0.2% YTD) are slightly negative through eight months. Gold was up 9.9% for the month but only 3.1% for the year, having peaked in January. Bitcoin gained a quarter of its value in August and is still down 10.7% on the year. Small caps had a down month and are still up 20.8%. All that to say, single months can be noisy.
Market Narratives
What’s Driving Interest Rates
Interest rates can be impacted by many factors including deficit spending and government debt issuance; monetary policy and a new Federal Reserve chairman; and inflation. The Wall Street Journal ran a nice short piece last week, going through some of these pressures. But the accompanying chart tells the story. Geopolitical conflicts (Iran) can cause both supply issues (oil, agricultural products, etc.) and deficit spending ($1.5 trillion defense budget request). I’ll touch on the spending and debt questions below, but inflation is the immediate issue.
To bottom line it: Higher energy feeds directly into inflation expectations, and sticky inflation is what keeps the Fed not only from cutting but potentially pushes it toward raising short-term interest rates. Then “higher for longer” becomes the bond traders’ mantra.
The 10-year U.S. Treasury Note yield is 4.79% as I write this, up from 3.97% since the conflict with Iran started (February 28th). The 10-year is widely viewed as a benchmark for the middle ground of the U.S. economy. It impacts mortgage rates, bond yields, and stock valuations. For reference, during this same stretch, the average 30-year mortgage rate has increased from 6.00% to 6.71%, and the core bond index has declined by 2.23%.
The practical takeaway is that intermediate and long rates are no longer a function of what the Fed does next. They are a function of inflation expectations, issuance, and who is willing to buy. That is a different regime from the one most of us were in from 2009 through 2021.
Going forward, we should expect greater emphasis on data now that markets can no longer lean on Fed guidance, raising uncertainty and the volatility risk premium around each meeting. September will be a live meeting, and avoiding a hike would likely require a downside inflation surprise. Intermediate rates appear anchored at these elevated levels for now. None of this should be read as risk-off. Fed Chair Warsh’s view of the economy, and of AI’s growth potential, is constructive, not alarmist.
For fixed-income investors, rising rates produced the slight bond declines in the chart above, but the same move bumped up the income side of the equation. Starting yield is the most reliable predictor of what a bond portfolio returns over the following decade, and today’s starting point is, again, constructive.
Forty Trillion Dollars
Total public debt crossed $40 trillion in August, roughly double where it stood a decade ago. Not great. The more concerning element is the cost of the debt at current interest rates, and acceleration in deficit spending during a solid and growing economy.
I should note that the headline tends to bring out a lot of doomsayers and pessimists. Ray Dalio and even Jamie Dimon have been making comments about points of no return and debt spirals for the better part of a decade now. A decade that had tremendous economic and stock market growth.
Dr. Ed Yardeni made a couple of points that don’t usually make the headlines. First, about $7.7 trillion of that total is intragovernmental or money the government owes itself for Social Security and federal retirement and healthcare obligations. It isn’t traded, so it doesn’t compete for buyers in the Treasury market.
Second, the 10-year yield is not greater than nominal GDP growth. Also, at 4.79%, the 10-year sits inside the 4-5% range that prevailed from the late 1990s through 2007. This could be simply a return to the “old normal” or pre-Great Financial Crisis growth and interest rate regime.
Our view is that the useful metric is interest expense as a share of revenue, not the headline balance, and the useful distinction is between “unsustainable” and “imminent.”
The naysayers may well be right eventually. Anyone who positioned a portfolio around that call has likely missed a large rally in the meantime. Both things are true at once. We can make tilts around long-term capital market assumptions, valuations, and factor exposures. What we won’t do is restructure a portfolio around a predicted crisis with no date attached.
AI’s Growth Contribution
How much is the artificial intelligence boom adding to U.S. economic growth?
J.P. Morgan Asset Management published a piece that puts numbers to the question everyone has been asking loosely. Hyperscalers (Microsoft, Amazon, Google, Meta, Oracle, etc.) are on pace to spend close to $800 billion on capital expenditures this year, roughly ten times their 2019 level. Investment in tech equipment and software, measured against GDP, has already passed its dot-com peak.
And yet, by their estimate, AI-related investment contributed only 0.47 of the 2.1% real GDP growth of the past year, or about one fifth of the total, after netting out imported hardware. Consumer spending contributed more than three times as much.
The gap between the spending and the measured growth comes down to three things.
- Not all hyperscaler capex is domestic (roughly 30% goes to international data centers) or new (these companies were already spending around $150 billion a year before AI).
- Imported hardware is subtracted in trade. And the U.S. design value in a chip like Nvidia’s, engineered here but fabricated and assembled in Asia, often never gets counted as a U.S. export at all. Epoch AI estimates that single issue has understated annualized growth by about 0.3 of a point.
- Some of the increases are price, not quantity. The price index for business investment in computers is up 10.9% year-over-year on surging memory prices.
On point 2, an AI server assembled in Asia still adds to the U.S. capital stock once it is running in a U.S. data center and hopefully gives a U.S. worker more capability and productivity, even though it technically didn’t add to U.S. real GDP growth.
Why does this matter beyond macroeconomic accounting? Fed Chair Warsh has put the productivity question on the Fed’s agenda and created a task force to study it. (There are also bigger questions, like whether AI is complementary to or competitive with our labor force.)
If the AI buildout is generating more real growth than the statistics capture, then policy is tighter than it looks. If it isn’t, then we have an enormous capital investment cycle sitting on top of a 1.5% GDP print.
It’s not an excuse for lower reported GDP numbers, but it’s worth remembering every time you see a GDP print this year described as disappointing.
Closing Thoughts
Let’s address the elephant in the room for September. And my typical disclaimer on seasonality: take it with a grain of salt. September has been the weakest month of the year for stocks in the post-WWII market, with both the lowest average return and the lowest odds of finishing positive. Midterm years tend to amplify it. We are in one.
That is a real pattern, and it is also only a pattern. Poor Septembers have mostly come when markets were already under pressure or the economy was already deteriorating. Neither describes the setup going in. Credit spreads are tight, volatility is low, earnings are growing at an unusually fast pace, and current estimates point to solid third-quarter growth.
There is also something unusual happening with earnings estimates themselves. In a normal year, analysts start out optimistic and spend the following quarters walking their forecasts down. Over the past two decades, the average calendar-year estimate finishes about 8% below where it started. This year has run the other direction. Results have come in ahead of expectations often enough that analysts have been raising estimates rather than cutting them, which is how the 2026 earnings growth number got to 31.2%. Whatever eventually knocks this market off course, it is not currently a downgrade cycle.
So, we are not repositioning anything around the calendar. What is worth watching is the September CPI report. Warsh has said the 2% target is not moving and that the Fed still has work to do, which leaves a hike on the table if inflation doesn’t cooperate. That print will tell us more about the next few months than any seasonal tendency.
As always, please reach out if you’d like to talk through how any of this affects your specific situation.
Thank you for reading.
