Thursday, July 23, 2026

Macro Musings · Daily Briefing · Thursday, July 23, 2026

Your energy transition got rerouted through a $100 barrel of oil and a 5% Treasury yield

Sulfur up 146% since start of the Iran war. Brent crude oil surged past $100/barrel as the Strait of Hormuz was largely closed due to US-Iran tensions, Houthi attacks in the Red Sea, and Black Sea disruptions, triggering a global energy price shock and inflation concerns.

Personal Stakes · Est. read time 5 min

In 30 seconds: Brent crude oil surged past $100/barrel as the Strait of Hormuz was largely closed due to US-Iran tensions, Houthi attacks in the Red Sea, and Black Sea disruptions, triggering a global energy price shock and inflation concerns. The US 10-year yield approached 4.70% and the 30-year held above 5%, with rate hike probability jumping to 36% as rising oil prices and ballooning deficits pressured bond markets and reignited Fed tightening debate. New Stripe Economics research found that while AI drives meaningful productivity gains at the worker and firm level, aggregate US productivity growth is being driven mainly by firms running existing capital harder rather than true AI-driven efficiency gains. Tesla and Alphabet both fell sharply after earnings—Tesla posted declining margins and net income while Alphabet reported negative free cash flow for the first time due to massive AI capex, dragging Consumer Discretionary and Communication Services sectors into oversold territory.

Brent settled at $99.80 on 2026-07-23, up 6.09% on the day, while WTI closed at $91.49, gaining 5.37%. The Strait of Hormuz has been largely closed for the better part of five months, and activity there has fallen back to near zero. Only 3 supertankers slipped through in the past 24 hours, hauling a combined 6 million barrels. That is not a functioning chokepoint. That is a trickle. The commodity shock extends well beyond crude. Since the start of the conflict, Sulfur is up 146%, European Natural Gas is up 94%, Heating Oil up 64%, Jet Fuel up 46%, Diesel up 39%, and Gasoline up 37%. Gasoline up 37% since start of the Iran war. Physical differentials are running $3–$6 per barrel above benchmark. Meanwhile the vast majority of asset classes are down, including stocks, bonds, crypto, and gold. Notably higher today is energy. $100 Brent crude complicates the economic and corporate picture, according to El-Erian. Brent crude: $100. One macro analyst warns that the entire rate complex will come unhinged whether the Fed hikes or holds. He is comfortable with air strikes and efforts to degrade Iran's capabilities in the Strait of Hormuz, but appears unwilling to significantly escalate against more vulnerable targets. US refineries are running at 96% utilization, processing 17.3 million barrels/day, and global oil inventories continue to decline. Global electricity demand growth is projected at 3.6% in 2026, alongside record global coal-fired electricity output. You wanted an energy transition. You got an energy crisis first.

The same dynamic driving Brent Crude Hits $100 as Hormuz Tensions Escalate is also a factor in US Bond Yields Surge; Fed Rate Hike Odds Rise.

The US 10-Year yield has pushed to 4.70%, its highest level of Trump's second term, while the US 30-year bond yield has held above 5% for the longest stretch since the dawn of the financial crisis. The market is now pricing in at least one Fed rate hike by December, with hike probability jumping from 12% to 36% over the past week. The hawks, per one Fed watcher, have a conviction that those inclined to wait may lack. In 1999-2000, the Fed hiked while crude ran from $9 to $35, a 300% move. From 2004 to 2006, it hiked 25 basis points 17 times over 17 meetings as crude climbed from $30 to $70, up 130%. The Fed has tightened through oil shocks before. Whether it should now is another question. Real Yields operate with cycles or regimes. During the 'money printing' era from 2009 to 2022, the 10-year real yield averaged just 0.23%. Before this, the 10-year real yield averaged 2.74%. US 10-Year yield: 4.70%.

Here is a puzzle. US labor productivity growth over the past year came in at 2.5%, well above the 1.6% 20-year average. The US is meaningfully above long-run trends, and models suggest it is currently in a high-productivity regime, though it's uncertain how long this will persist. That sounds like a technology story. Academic research is coalescing around the finding that AI drives meaningful productivity gains at the micro-level: tasks, workers, firms, and these findings are particularly compelling because they largely use older generations of models, with newer models likely even stronger. So far so good. The problem is that these micro gains are not showing up meaningfully and persistently at the aggregate level in current data. If AI were driving macroproductivity gains, we'd expect to see an acceleration in total factor productivity (TFP), but recent estimates from both the San Francisco Fed and BLS have been weak. San Francisco Fed estimates suggest the 2024 aggregate US labor productivity gains were not TFP-driven, but 2025+ gains were due to greater capital utilization: firms running their existing capital hotter (e.g. more GPU cycles). Firms are pushing the limits of their existing infrastructure plausibly to meet the demand for AI capacity. The productivity boom, in other words, is less about a new way of working and more about squeezing existing infrastructure harder. Why the disconnect? AI is speeding up certain tasks, but firms haven't yet adjusted to ease downstream bottlenecks and unlock these gains in aggregate. The task gets faster, but the workflow around it has not been redesigned to absorb the speed. And at the industry level, recent AI adoption weakly predicts industry-level productivity growth over the last 3 years, but also weakly predicts pre-pandemic productivity growth (which obviously wasn't driven by AI), and after controlling for pre-pandemic growth, there is no correlation. The correlation vanishes once you account for which industries were already productive before anyone had heard of a large language model. The microproductivity research remains promising, especially as it shifts to newer models, and the bottleneck story is exactly what you would expect in early adoption. The revolution, if it comes, has not yet arrived in the data that counts.

Earnings season has a way of reminding you that even the largest companies in the world are, at the end of the day, just companies. They report numbers, and sometimes the numbers are bad, and then the stock goes down. VIX up 12.68% on the day.

Personal Stakes · personalstakes.com unsubscribe · manage preferences