With the U.S. 10-year Treasury yield above 5%, the market is asking a different question now. The focus is no longer just whether the Federal Reserve will keep raising rates, but what it would take for higher borrowing costs to actually change behavior across the economy.
In the latest Weekly Roundup from Forward Guidance, DCP, a trader with about 40 years of experience in rates and fixed income, said the real signal for a bond-market reversal may not come from a round-number yield target such as 5.5% or 6% on the 10-year. He argued it is more likely to come when some part of the credit system or capital-spending cycle starts to break.
The market is looking for the rate level the economy can actually bear
The MarsBit article says the debate has shifted as Treasury yields keep rising and markets price in more tightening. Once the 10-year moved above 5% and the front end swung from rate-cut expectations to pricing multiple hikes, the bigger issue became how high rates must go before they can truly restrain capital spending.
DCP said this move higher in yields is no longer only a repricing of the Fed path. In his view, the market is searching for a rate level high enough to constrain investment. For years, investors treated rate hikes as the opening stage of slower demand, cooler growth, and eventually rate cuts. This time, large technology companies are still spending on data centers, compute, power, and other AI infrastructure. Higher rates are trying to suppress demand, but a large capex engine is still running on the other side.
That is why, according to DCP, yields may struggle to find a durable path lower as long as AI investment does not slow in a visible way. He said the bond market now needs to watch not just a single Federal Open Market Committee meeting, but whether financing costs rise enough to make marginal AI projects uneconomic.
The pain trade may not be over
Around the time the program was recorded, the 10-year Treasury yield briefly touched about 5.23%, the highest level in nearly 19 years, while the 30-year yield moved above 5.5%. At the same time, expectations for further Fed tightening strengthened. Reuters reported then that rate markets had priced in as much as about 90 basis points of additional tightening.
This was not a routine move in yields. Over the previous two months, the long end of the Treasury market had sold off sharply even though markets had earlier been trading rate cuts. DCP said on the show that the front end had moved quickly from pricing cuts to pricing multiple hikes.
That left a simple question: how much further do bonds need to fall before buyers step in? DCP’s answer was that there was still not enough reason to rush into the market. In his view, a 5.2% 10-year yield might still be insufficient for a durable bottom.
What DCP wants to see before turning constructive on bonds
DCP laid out a clear list of conditions that could support a more bullish bond view. He said the market likely needs one or more of the following to change in a meaningful way:
- the Fed signaling that the hiking cycle is close to ending;
- a visible slowdown in AI capital spending;
- easing pressure in refining margins, diesel, and energy prices;
- a meaningful correction in equities;
- a clear deterioration in the labor market, or another shock large enough to force a repricing of the policy path.
Before that happens, he still sees higher yields as the pain trade.
The article stresses that the key point is not a precise yield target. It is DCP’s view of what this rate cycle represents. Under normal conditions, sufficiently high rates lift financing costs, reduce demand, cool the economy, and eventually remove the need for more tightening. This cycle looks different because of AI. Large technology companies are still making heavy infrastructure investments, and spending on data centers, power, chips, and related facilities is supporting both corporate investment and, to some extent, broader U.S. economic activity.
That means higher rates are trying to cool demand while AI investment continues to create capital demand. DCP even suggested that the market may need higher real rates than previously assumed to create a meaningful constraint. He framed that as a market view, not a proven equilibrium. On the program, he said the 30-year Treasury yield may need to move above 6%, with short-end policy rates also rising further, before capital allocation behavior changes in a real way.
The problem, he said, is that by the time rates are high enough to restrain AI investment, they may already have damaged more fragile parts of the economy first.
The Fed can suppress demand, but it cannot solve every inflation driver
DCP said the policy backdrop is especially difficult because this inflation cycle is not being driven entirely by classic demand overheating. Energy, diesel, transport, and agricultural costs are still affected by supply-side forces, and monetary policy has limited power over that kind of shock.
The article says the Fed raised its federal funds target range by 25 basis points to 3.75%-4.00% on Sept. 16, the first rate increase in three years, with the FOMC approving the move by a 12-0 vote.
But inflation pressure was not coming only from demand. Energy prices were still being pushed around by geopolitics and supply disruptions. By late September, the continuing conflict in the Middle East was still lifting energy costs and had become one of the main backdrops for the renewed rise in global bond yields.
That creates a policy problem. Rate hikes can reduce demand for consumption, housing, and corporate financing, but they cannot produce more oil or directly remove an energy supply shock. DCP summed it up this way: the bigger concern is not what the Fed can solve, but what it cannot solve.
If inflation comes from strong household demand, the Fed can lean against it by raising rates and slowing consumption and credit. If inflation is driven mainly by an energy supply shock, rate hikes do more to limit pass-through into other goods and services than to fix the shock itself. In that setup, if oil and diesel prices stay elevated, the Fed could face an uncomfortable mix in which more sectors feel the pain of high rates while inflation still does not fall enough to make policymakers comfortable.
AI is supporting growth, but it could also become the turning point
One of the most closely watched parts of the discussion was DCP’s view that AI sits in two places at once in this cycle: as a support for the expansion and as a possible breaking point.
Large technology companies are generating major financing and investment demand through capex. That is contributing to a more visible split in the U.S. economy. AI, data centers, and investment tied to large tech companies remain strong, while pressure keeps building in housing, small businesses, and other sectors that are more sensitive to financing costs.
As of late September, even with long-dated Treasury yields at multi-year highs, global equity markets had remained fairly resilient, and enthusiasm around AI investment was one reason. For that same reason, DCP said AI may be the most important variable to watch if investors are looking for what could reverse the direction of rates.
His logic was straightforward. AI infrastructure requires large amounts of capital. If bond yields and corporate financing costs keep rising, the hurdle rate for new projects rises with them. A company deciding whether to build a data center at a 6% or 7% funding cost is making a very different decision from one facing a 9% or even 10% cost of capital.
In other words, the rates market may not end up “beating AI” directly. It may instead keep lifting the cost of AI expansion until marginal projects lose their economic appeal.
DCP therefore highlighted private credit and corporate credit markets. If a large AI-linked company, a private credit fund, or a highly leveraged project runs into financing trouble first, that could trigger a broader reassessment of the entire AI capex cycle.
The article notes that this remains a risk scenario, not an event that has already happened. Still, it points to a market mechanism worth watching. Over the past few years, rising U.S. government issuance has already increased fixed-income supply. If AI companies also enter bond and private credit markets at scale to raise funds, the public and private sectors could end up competing for the same pool of capital.
That would push up the price of money and create a potentially self-reinforcing loop: more financing demand leads to higher yields, higher yields lead to higher capital costs, and higher capital costs make the weakest projects more likely to fail first. If that chain breaks, it could become the real buying window for bonds.
The first casualties may not be large-cap tech
DCP also argued that headline indexes are masking how unevenly high rates are being transmitted through the economy and the stock market.
On the program, he listed companies such as Nike, PayPal, Disney, Pfizer, Home Depot, and Lululemon, all of which had fallen sharply from their historical highs. His point was not to recommend those stocks. It was to show that an index staying elevated does not mean the whole market is in a bull run.
Large technology and AI companies now carry so much weight in major indexes that gains in a small group of names can hide weakness across a much broader set of companies. That led the guests to ask a pointed question: if you strip out the Magnificent 7 and the main AI beneficiaries, where is the “real market”?
DCP said the split resembles the K-shaped structure seen in the real economy. Cash-rich large technology companies can absorb higher financing costs and may not need outside funding at all. Small businesses, commercial real estate, and companies that must keep rolling debt are operating in a very different rate environment.
Using U.S. small-business financing as an example, he said some loan products are already carrying double-digit borrowing costs. In that environment, the key question may not be whether Apple or Microsoft can keep investing. It may be when smaller companies with weaker cash flow and ongoing refinancing needs start to show more restructurings and bankruptcies.
That is why he said investors should watch small-business bankruptcies, private credit, commercial real estate, and regional banks. If long-term yields keep rising, those areas may reveal the pressure from high rates earlier than the major equity indexes do.
Would 6% be the real breaking point?
The market is paying more attention to one number: 6%. DCP said the U.S. economy would not automatically fall into crisis just because the 10-year Treasury briefly moved above 6%.
What matters more, he said, is the speed of the move and whether markets can find a new equilibrium at higher yields. If yields rise gradually, investors may slowly accept the new pricing. Pension funds, insurers, endowments, and individual investors would then face a more practical question: if risk-free assets offer sufficiently high returns, why take so much equity and credit risk?
DCP also said that once long-dated Treasury yields become attractive enough, asset-allocation money will naturally begin to move in. The issue is timing. When do yields become high enough to attract buyers, and when do they become high enough to damage the economy first?
That gap is the balance point the bond market is still trying to find. Recent price action suggests it has not found it yet. On Sept. 29, the U.S. 10-year Treasury yield rose further to 5.293%, while the 30-year reached 5.6206%, levels last seen in 2007 and 2002, respectively.
In other words, even if a 5.22% 10-year yield looked extreme when Forward Guidance recorded the episode, the market still pushed to higher yields only days later.
Four variables to watch for a real bond-market bottom
If DCP’s hour-long discussion is reduced to a framework, the next step is not deciding whether 5.2% is high enough. It is watching four sets of variables.
- Whether Fed hiking expectations have truly peaked. After the September rate increase, policymakers still had not sent a unified signal that tightening was over. New York Fed President Williams later said there was no need to rush into another hike, but still viewed one more increase this year as reasonable. If inflation cools and rate markets begin to consistently price out further hikes, that would be the first major condition for a shift in bonds.
- Whether energy pressure eases. If crude oil, diesel, and related costs keep falling, the inflation impulse from supply shocks would weaken and the case for more tightening would also fade.
- Whether AI capex starts to slow. As long as large technology companies keep investing aggressively in data centers, power, and compute, high rates may struggle to bring down overall investment demand. If financing costs begin to affect AI investment decisions in a visible way, or if credit markets crack first, the bond-market narrative could change quickly.
- Whether the financial system shows a real stress point. Private credit, commercial real estate, regional banks, small-business bankruptcies, and credit spreads may reflect the damage from high rates earlier than stock indexes do.
That is also why DCP said he is still unwilling to buy bonds too early. The market has already moved from asking when rate cuts will begin to asking how many more hikes may still come. But the thing that could reverse the trade is not a round number on a yield chart. It is a real-world constraint finally starting to bite.
Put another way, the end of this Treasury selloff may not depend on whether the 10-year reaches 5.5%, 6%, or something higher. It may depend on whether high rates finally force AI investment, credit expansion, energy-driven inflation, or some part of the real economy to give way first. Until then, the level of yields is still the result. The more important question is what breaks first.