The Takeaway

The mid-August Consumer Price Index report landed about where the Fed wanted—inflation cooling, though slightly, not climbing. Paired with the unexpected loss of 23,000 jobs in July, the Fed has the cover it needs to keep rates steady at its next meeting on Sept. 16. A full month of data still awaits, but right now that path seems clear.

While there had been broad expectations of several cuts in 2026, now it’s quite the opposite. The current odds favor at least one or two increases over the next year.

The logic is simple: the Fed funds rate will rise if inflation flares. If that happens, a one-time, quarter-point nudge likely wouldn’t be enough. A single hike could become two or more.

The bigger wildcard may be the Fed’s tone, not its rates. New Fed chair Kevin Warsh believes the bank had become too talkative—that the pendulum had swung too far toward more communication and guidance—costing it some flexibility. Expect less foreshadowing from here on. That doesn’t mean silence, just room for surprises. When the Fed surprises, markets typically get bumpier.

Warsh is still new to the job, so whether he holds that line remains to be seen. For now, though, the data says to wait. And the Fed will, most likely.

At the most fundamental level, interest rate hedging is simply locking in today’s expectations for tomorrow’s interest rates. But this year, those expectations flipped—from rate cuts to probable rate hikes—faster than anyone expected. The whiplash drove up hedging costs and reduced hedging activity. As protection became pricier, fewer customers bought it.

That hesitation alone carries a price. Every month spent waiting, as expectations for higher rates climbed, has only made the eventual hedge more expensive. Customers who paused are now understandably second-guessing their decision.

The market isn’t offering much relief. Nearly two rate hikes are now priced in over the next year. For anyone still on the fence, the past few months have delivered a blunt lesson: waiting can have a cost.

The tightest credit spreads in the primary lending market since 1997 are a reassuring sign. Tight spreads typically indicate a stable economic outlook and lower perceived risk.

The catch is what lies beneath. With base and Treasury rates staying higher for longer, all-in yields remain elevated even as margins compress. Cheap spreads, costly money.

The glaring exceptions are the hyperscalers, bound and determined to execute their AI data center plans in hopes of leading the AI revolution. Fueled by more than $800 billion in AI capital spending this year, their borrowing needs have surged, pushing their spreads from Treasury-plus-50 to Treasury-plus-100.

For now, the widening is limited to the hyperscalers, but funding supply and demand could cause it to spill over into other sectors.

One quieter shift: with SOFR below most Treasury rates, floating-rate loans look more attractive than they have in a while. As a result, lenders expect a busy September and October, driven by infrastructure construction and agribusiness M&A.

The secondary loan market, meanwhile, has split in two. Higher-rated paper has held steady—double-B spreads have barely budged, moving 17 to 19 basis points all year—while the bottom of the market buckled.

The single-B-minus tier swung nearly 130 basis points as retail outflows from loan funds, driven by earlier expectations of Fed rate cuts, forced selling amid softening CLO demand. CLOs remain quality-focused, and investors continue to play favorites.

Among sectors, semiconductors, construction and engineering are thriving. Software, affected by AI disruption, and fuel-reliant trucking are lagging.

Still, in the bigger picture, this anomaly seems moderate—double-B spreads remain well within their three-year averages.

For U.S. agribusiness, the 2025 crop marketing year is turning out reasonably well, at least better than many had feared. Strong yields and occasional price rallies are making a few operations moderately profitable, while most are breaking even. The weak spot is agronomy, where excessive competition for limited demand is reducing revenue and squeezing margins in many regions.

At this point, the 2026 crop looks more uneven. The Eastern Corn Belt is strong, while the Dakotas are struggling with poor yield forecasts. USDA recently adjusted its estimate of carryout—the grain left in storage when the marketing year ends—to just above 10%. Typically, when that cushion falls below 10%, supplies tighten and prices can swing sharply, so discipline and flexibility matter when bidding for grain.

Globally, news headlines are creating more volatility than providing direction. Attacks on grain ships in the Black Sea and ongoing tensions in the Middle East are making headlines, but prices tend to spike, then settle. In many respects, the old expression “Buy the rumor, sell the facts” has never been truer.

The bigger pressure remains on input costs—higher fertilizer and sulfur prices, European buyers outbidding U.S. buyers for fertilizer, elevated fuel costs and now the prospect of rate increases have operating margins under pressure.

All of this is prompting interest in consolidation, some by necessity and some by opportunity. Combining operations can make better use of existing assets while preserving the service farmers depend on and, in many cases, preserve the grower’s most important partner, the local co-op.

During its early-August earnings call, SpaceX outlined its mobile-telephone ambitions: it aims to compete with AT&T, Verizon and T-Mobile by using a network built from its thousands of satellites and small ground stations, rather than relying on traditional cell towers. The company’s Starlink subsidiary recently purchased 65 MHz of spectrum from EchoStar—supplementing its existing 5 MHz—to support that goal.

Starlink’s combined 70 MHz of spectrum pales in comparison to the 300 MHz each of the big three providers uses. But SpaceX founder and CEO Elon Musk has a history of doing what others consider impossible. And Musk’s eye is likely set on more than just phone plans—it’s on the bundle.

Starlink already sells home broadband to about 12 million subscribers, 85% of whom are in rural America, according to Recon Analytics. Pair that with mobile service, and it becomes a compelling offer, especially as satellite internet continues to become cheaper.

For rural internet and mobile providers, that’s a potential headwind to watch. It won’t happen overnight, and it’s no cause for alarm. But cheaper, better and everywhere is a tough combination to beat.

Artificial intelligence runs on electricity, and the bill is landing in unexpected places.

As onshoring of manufacturing increases and data centers multiply, utilities are rushing back to natural gas for round-the-clock power. But as Teri Viswanath, lead economist, Power and Energy at CoBank, details in her report Dash to Gas, this build cycle will be slower, costlier and more inflationary than past cycles, largely because turbines, transformers and interconnections are in short supply.

Meanwhile, demand is staggering. A Barclays analysis found that Google’s electricity use jumped 37% last year to more than 43 terawatt-hours. Amazon’s data centers drew roughly 90 terawatt-hours. Efficiency and renewables help, but they don’t address transmission and firm capacity.

Co-ops and other utilities—as well as public utilities commissions—remain hyperfocused on affordability. Structured agreements and demand-response deals to allocate energy costs appropriately are now commonplace. In recognition of their role, hyperscalers have signed pledges to pay their own way and not shift these costs on to consumers and small businesses.

The information provided in this publication is for informational purposes only and is not to be used or considered as investment research, a proposal, or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. Companies and transactions referenced in this publication are shown for illustrative purposes only, and the provision of such information is not a recommendation or endorsement in this context. Certain information contained in this publication has been obtained or derived from third-party sources, and such information is believed to be correct and reliable but has not been independently verified. While CoBank believes that factual statements in this publication, and any assumptions on which information in this publication is based, are in each case accurate, CoBank makes no representation or warranty regarding such accuracy and shall not be responsible for any inaccuracy in such statements or assumptions. Note that CoBank may have issued, and may in the future issue, other reports that are inconsistent with or that reach conclusions different from the information set forth in this publication. CoBank is under no obligation to ensure that such other reports are brought to your attention. Furthermore, the information may not be current due to, among other things, changes in the financial markets or economic environment, and CoBank has no obligation to update any such information contained in this publication. This publication is not intended to forecast or predict future events.