Over the past week, global markets were once again driven by interest rates. The U.S. 10-year Treasury yield touched 5.36% at one point, while the 30-year yield rose to its highest level since 2002; the 10-year Treasury auction priced at the highest yield since 2000, but near-record overseas demand pulled sentiment back. Across the Atlantic, the France-Germany government bond spread widened to 140 basis points, returning to the stress range seen during the 2011-2012 European debt crisis.
Bank of America strategist Michael Hartnett compressed his core judgment into three words in the latest Flow Show: "Bond bubble ends." The most counterintuitive inference followed: the most bullish event for risk assets before year-end would be a Fed rate hike in October 鈥?using a proactive tightening to stop long-end yields from spiraling further out of control.
This logic is not contradictory. U.S. stocks are still up 13.4% year-to-date, with the S&P 500, Nasdaq and global tech indices still at high levels, but government bonds have fallen 4.4% year-to-date, lagging cash returns by 7.2 percentage points. Stocks have not yet been knocked down by rates because U.S. hard data, global earnings expectations and hyperscaler capex expectations have not been significantly revised lower; bonds are the first to create cracks in more fragile places: Korean brokerages, French government bonds, and CCC-rated credit.
The real risk is not just "high rates," but that after rates rise, fiscal policy, elections, AI capex and fund flows are all forced to reprice. If the Fed does not proactively hike, long-end yields may continue to push higher; if Democrats sweep both chambers in the midterms, the political discount rate for stocks will suddenly change; if cash yields remain above 4%, the $8 trillion in money market funds may not be in a hurry to enter the stock market.
What bonds hit first is the part most dependent on cheap money
Hartnett used South Korea as a sample. Over the past 18 months, Korean bond yields rose by about 200 basis points; Mirae Asset, South Korea's largest retail brokerage, fell 65% since May after previously becoming a "four-bagger."
This does not mean the Korean stock market bubble has already burst. Korean stocks are still up 98.9% year-to-date, one of the strongest-performing markets globally. It shows something else: when the cost of funds suddenly rises, the first casualties are often the areas with the highest financial leverage and trading heat.
Global central banks are also accommodating this shift. Since 2026, global central banks have raised rates 54 times; this is the first time since September 2023 that global central banks have turned back toward tightening on a net basis.
The divergence in asset performance further confirms this fracture. In the year-to-date performance ranking, oil is up 58.5%, U.S. stocks are up 13.4%, and global stocks are up 10.6%; cash returns 2.8%, government bonds fell 4.4%, gold fell 4.8%, and Bitcoin fell 6.7%. The "humiliation" of bonds is most striking: even holding cash has outperformed holding government bonds.
Why an October hike is viewed bullishly: not hiking could let the long end run out of control
The main drivers of rising U.S. rates are split into two parts: growth remains strong, and uncertainty is rising. Supply, commodities and fiscal concerns also have an impact, but they are not the core explanation.
The key change occurred after Jackson Hole. Previously, the market still had an anchor: policy was already restrictive, and rates could eventually return to about 3%. BofA rates strategists pointed out that after this "anchor line" was cut, yields would continue to rise until data worsen or the market itself runs into trouble. Their conclusion is not friendly to those waiting for a top: "We believe neither condition is currently met. Any dovish suppression by the Fed on short-end pricing will only shift rate pressure to the longer end."
This time, the rise in 10-year Treasury yields since late September has mainly come from term premium, not from a repricing of the fed funds path. That makes the problem more difficult.
Hence the seemingly strange judgment: the Fed either hikes, or faces continued spike-ization of bond yields. The September FOMC minutes showed that all 19 officials supported a hike, and most thought one more hike this year would be appropriate. But the market currently assigns only a 17% probability of a hike at the October 28 meeting, and an 81% probability for December 9. BofA rates strategists believe market pricing is wrong and have established corresponding trading positions (paying October FOMC OIS). Their wording is concise and direct: "If U.S. data remain strong, the Fed must choose: hike, or tolerate bond yield steepening. If frightened by bond steepening, the Fed will hike."
The bullish logic here is not that "a hike itself is good for stocks," but that if the Fed uses an October hike to prove it is willing to control inflation and term premium, the market may instead stop worrying about the 10-year yield heading toward 6%, and pressure on stocks would ease accordingly.
Fiscal cracks have already appeared, and France is the clearest example
The other half of the rising-rate problem is who pays the higher interest. BofA chief global economist Claudio Irigoyen pointed out that the rate shock is "exposing the unsustainable reality of fiscal policy in developed economies," with the U.S. and France first in line, followed by Japan and the U.K.
The root of the problem lies in what economists call time inconsistency. Over the past 20 years, U.S. debt/GDP rose by about 60 percentage points, and France by about 50 percentage points, most of it during the zero-rate era. This debt supported permanent spending and permanent tax cuts. Now financing costs are normalizing, but cutting spending or raising taxes both lack political votes. Both countries have debt/GDP at about 120%, and primary deficits above 2.5% of GDP; the U.S. overall deficit is about 6% of GDP, while France's is about 5%, with the U.S. higher mainly because its interest burden is heavier.
France is the clearest rupture. According to Bloomberg, the France-Germany government bond spread has returned to 140 basis points, the fastest selloff in French sovereign spreads in at least 20 years, even faster than in 2011. The market has no patience for French fiscal policy, and several data points make that clear: since the post-pandemic peak, developed economies' debt/GDP has fallen by an average of 11 percentage points, but France has not declined; France's private non-financial debt reaches 214% of GDP, versus 140% in the U.S.; the median net income of French retirees is almost close to that of active workers; and France has not achieved a primary budget surplus since 1974.
According to The Wall Street Journal, Paris is considering increasing short-term government bond issuance. Rabobank commented that this is "the latest example of a developed-market economy behaving like an emerging market." BofA European credit strategist Barnaby Martin expects the France-Germany spread to fluctuate in the 100 to 150 basis point range; if it rapidly approaches the upper end of that range, the probability of short-term debt issuance, verbal ECB intervention, and even discussion of the TPI tool will rise.
Stocks have not fallen yet, but that does not mean rate hikes have not bitten
High rates have not been without impact; they just have not yet bitten the market's most core assets.
U.S. hard data have not clearly rolled over, and global forward earnings expectations have not rolled over either. More importantly, hyperscaler capex expectations are still rising, and 2027 consensus expectations are still climbing steeply. As long as AI capex continues to support real demand, there is reason for yields to keep rising. BofA strategist Barnaby Martin's judgment is this: the moment U.S. macro data begin to roll over is when rate volatility peaks and the credit market gets a green light again.
The credit market is already reflecting pressure at the edges. In the September widening of high-yield bond spreads, less than 30% could be explained by higher refinancing costs or expected losses, with the rest looking more like a rate volatility premium. CCC spreads have risen to 1,200 basis points, the highest since 2022.
Stocks are not entirely without cracks internally either. The BofA Bull & Bear Indicator fell from 8.8 to 8.1, after being 9.3 two weeks ago, still in "sell" territory. The Global Breadth Rule shows that a net 50% of global stock indices are oversold, meaning they are below both their 50-day and 200-day moving averages; but a contrarian buy signal is not triggered until a net 88% threshold, and unless tech-heavy markets such as the U.S. and South Korea reverse quickly, that threshold will be hard to reach in the short term.
$8 trillion in cash is not a natural buyer
The fund flow data send a fairly clear signal.
As of Wednesday of that week, cash inflows reached $166.4 billion, the largest since April 2020; bond inflows were $33.8 billion, stock inflows $12.4 billion, gold inflows $2 billion, and crypto asset outflows $0.6 billion. The $118 billion in cash outflows the previous week due to quarter-end factors not only all came back, but flowed in beyond that.
By segment, Treasury inflows were $16.2 billion, the largest since April 2025; investment-grade bond inflows were $10.3 billion; bank loan inflows were $2.4 billion, the largest since February 2025, with floating-rate products becoming a hedge against rate hikes. Tech stocks still saw $3.5 billion in inflows, while financial stocks saw $3 billion in outflows, the largest since March 2023.
Money market fund assets have reached about $8 trillion, up from $5 trillion in 2023. At a 4.2% T-bill yield, this money generates about $330 billion in annual interest income.
Hartnett's liquidity signal is even more direct: "No rate cuts, no decline in cash." Historically, large-scale movement of sidelined cash into stocks usually happens after massive monetary easing, such as QE in 2009, QE4 in 2020, or during a sustained Fed rate-cutting cycle. The current environment does not meet that condition.
The midterm elections give the market an uncomfortable probability distribution
The U.S. midterm elections on November 3 are the event most likely to trigger 10% level stock market volatility in the coming months.
Polymarket currently gives Democrats a 64% probability of sweeping both chambers, up from 49% in September; the probability of Republicans holding the Senate and Democrats taking the House is 28%; and the probability of a Republican sweep is only 9%. Trump's approval rating is 38%, his approval on economic issues is 34%, and on inflation it is 27%.
The most uncomfortable part for the market is that the currently most likely outcome corresponds precisely to the most risk-averse path.
In a Democratic sweep scenario, stocks could fall more than 10%, the dollar and bond yields could decline, international stocks could outperform U.S. stocks, and Europe could outperform Asia. Tax and regulatory direction would move upward, which is negative for EPS. More importantly, the political capital for the Trump administration to push resource monopolies and corporate policy cooperation would decline.
If Republicans hold the Senate and Democrats take the House, that corresponds to a "divided government = Goldilocks" risk-on path, and stocks could rise more than 5%. If Republicans sweep, that is the biggest upside surprise, with stocks potentially rising more than 10%, and the AI trade continuing to advance toward a 2027 bubble.
The most critical state to watch may be Texas. Republican incumbent Governor Greg Abbott leads Democratic challenger Hinojosa by only 48% to 46% in polls. Texas is also a focal point for AI data center approvals and power strain, and the conflict between "affordability" and AI capex will be repriced in the ballot box.
Outside the U.S., the political right-turn trade continues
U.S. political risk skews negative, but overseas there is another line of logic: the right-turn trade.
Brazil's Bolsonaro won the first round, with the second round on October 25. Among 20 Latin American countries, 11, with about 420 million people, are under right-wing leadership; before the previous election round, that number was 7 countries and about 60 million people. The market reaction was direct: Brazilian stocks hit a record high, and the 10-year yield plunged 124 basis points in a single day.
The MSCI Latin America index is still 35% below its all-time high, which is also the reason Latin America is being included in the global rotational bull market sequence: the first half of 2025 was Europe, the second half of 2025 to the first half of 2026 was Japan and emerging Asia, and now it is Latin America and broader emerging markets' turn.
Europe also has similar political variables. Italy and Germany have already shown signs of a right turn, and Le Pen leads in polling for France's April 2027 presidential election. Spanish Prime Minister S谩nchez announced an early election, with voting set for November 29, and polls show the center-right leading 鈥?Hartnett characterized it as "an underappreciated positive event for European bonds."
Next, watch the price lines: several key levels matter more than the macro narrative
Under this framework, short-term positioning leans toward risk reduction before the midterm elections. Three clues support this judgment: global central banks are tightening net again, financial conditions are getting tighter, and early signals of credit events are still appearing, including AUD/JPY, risk parity strategies, and credit default swaps of tech companies such as Oracle and SoftBank.
Tech stocks are not suitable for continued chasing, because investors are already heavily positioned; MAGS may still outperform SOX, and Chinese tech stocks have contrarian appeal. On the bond side, position early for peak-yield trades: 30-year Treasuries, small caps, REITs, and Hong Kong property. The broad direction remains holding gold, commodities and emerging markets.
Several price levels are more operationally meaningful than macro slogans: 6% on the 10-year Treasury is the key area where bond buyers are waiting; a France-Germany spread of 150 basis points is close to the point where the probability of policy backstop rises; a net 88% oversold reading in the Global Breadth Rule is the stronger contrarian buy signal; a Bull & Bear Indicator below 8 is closer to a risk-easing zone than the current 8.1; and the 30-year Treasury auction yield has already reached 5.618%, which is also a practical anchor for positioning early in long bonds.
The calendar is equally dense: Brazil's second round on October 25, FOMC on October 28, U.S. midterm elections on November 3, Shenzhen APEC on November 18, Spain's general election on November 29, FOMC on December 9, and Miami G20 on December 14.
Hartnett concluded with one sentence: "Most investors are waiting for the 10-year yield to reach 6% before buying Treasuries; we are positioning modestly early in 'peak yield' trades." If the Fed hikes in October and suppresses the long end, risk assets may instead breathe a sigh of relief; if the midterms deliver a Democratic sweep, U.S. stocks will need to be repriced; and if yields continue to rise while data still do not roll over, the bond bubble-ending logic will continue to operate in the most fragile places.
Comments