Bank of America’s Hartnett: Japanese bank stocks are a “leading indicator of global risk-off sentiment,” with four major contrarian trading opportunities in the second half of the year

Bank of America believes the current market consensus is exhibiting a rare, broadly bullish posture—no one is bearish on the second half of the year. It warns that this consensus itself is what breeds the most值得关注的 contrarian trading opportunities.

Bank of America’s chief investment strategist Michael Hartnett outlined the market’s current “four nos” consensus in his latest report: no landing, no rate hikes, no cut to AI capital expenditure, and Democrats not sweeping the midterm elections.

He believes that any failure of the above expectations would deal a major blow to asset prices, and he accordingly proposes four categories of contrarian trading strategies.

Meanwhile, he issued a macro risk warning: once Japanese daily bond yields keep climbing and cause Japanese bank stocks to weaken, it will be a “warning signal” of a “global-scale risk-off sentiment.”

The market consensus’ four “nos” support risk appetite

Hartnett summarized the prevailing logic in the market as four layers of consensus, noting that these four layers together suppress the bearish forces for the second half of the year.

First, the U.S. economy will not avoid landing. The market broadly expects the economy in the second half will not slow down materially; nominal growth will continue, driving the allocation logic of “hold everything, except bonds”—“you can’t buy bonds, and you can’t sell stocks.”

Second, the Fed will not hike rates. Especially ahead of the midterm elections, rate hikes are unfavorable for both asset prices and the macro environment, so the Fed and major global central banks maintain a “moderately hawkish” stance.

It is worth noting that in this year of high inflation, the number of global central banks’ rate cuts in 2026 (34 times) still exceeds the number of rate hikes (21 times).

(the gap between rate hikes and rate cuts by central banks in developed countries and emerging markets globally; 3-month rolling statistics)

Third, AI capital expenditure will not be cut. The market’s consistent expectation is that super-large cloud providers’ AI capital expenditure will be about $800 billion in 2026, and will surpass $1 trillion in 2027.

Fourth, Democrats will not sweep the midterm elections. Bank of America’s June fund manager survey (FMS) shows that global investors assign only a one-fourth probability to Democrats sweeping both the House and the Senate, even though the probability in the prediction market Polymarket has risen to 45%.

(Trump’s approval rating)

Four contrarian trades, betting on the consensus failing

Hartnett believes everyone is pricing in a “no landing” scenario; once any consensus is broken, the market will face dramatic repricing. He therefore proposes four contrarian trading strategies.

Bet on “landing.” With everyone positioning for the economy not to slow, all it takes is for another round of weak nonfarm payrolls data to be enough to push the previously ignored long-duration bonds (10-year Treasuries), defensive sectors (such as consumer staples and the tech megacap “seven giants”—the latter are essentially defensive monopoly enterprises), and high-dividend stocks higher.

Bet on “rate hikes.” Hartnett points out that the best corresponding trade for rate hikes by the Fed ahead of the midterm elections is going long the U.S. dollar and going long for the flattening of the yield curve (i.e., betting on curve inversion).

He highlights a historically rare signal: U.S. CPI (4.2%) and the unemployment rate (4.2%) are close to matching—something that has only occurred over the past century in 1966, 1973, 1990, 2000, 2008, and 2021. And in each of those years, it ended with the Fed hiking rates—and none of them has ever been viewed on Wall Street as a “good year.”

Bank of America projects that from now through the end of this year, major central banks around the world are expected to collectively carry out 18 rate hikes and 9 rate cuts.

Bet on “AI capital expenditure cuts.” This is the most market-shocking surprise scenario relative to expectations. The corresponding trades are going long software and the tech seven giants, and shorting the Philadelphia semiconductor index.

Hartnett believes the trigger is that: super-large cloud providers’ cash flows remain persistently negative; their debt scale has already exceeded $208 billion; and the “bond vigilantes” may force them to issue stocks, cut hiring to sustain spending, and are thereby forced to compress capital expenditure.

Bet on “Democrats sweeping.” Trump’s supportive inflation rate keeps making the approval rating worryingly slide lower. If the Republican Party loses control of the Senate, it would create an event shock of “yields down, the dollar down, stocks down.”

Hartnett suggests that if after early September’s U.S. Labor Day, Trump’s approval rating does not show a clear rebound, investors should buy gold in September to hedge the top-end risks piled up by Wall Street’s “greed” sentiment.

Japanese bank stocks, the “canary” for global risk

On the macro big-picture level, Hartnett views Japanese bank stocks as a key indicator to monitor global risk appetite.

Over the past three years, as Japanese government bond yields surged from 0.5% to 3%, Japanese bank stocks have gained a cumulative threefold.

(Japan 10-year government bond yields and the TSE Banks index)

Hartnett notes that global bank stocks are broadly breaking upward, because this sector is one of the industries with the highest adoption of AI technology, and because policymakers’ stances across countries are relatively moderate.

However, he warns that once Japanese government bond yields rise further and instead cause Japanese bank stocks to weaken, it will become a “canary”-style early warning signal for the outbreak of “global-scale risk-off sentiment.”

“25/25/25/25” portfolio, annualized 16% return in-year

On the asset allocation level, the all-asset allocation plan proposed by Bank of America strategist Hartnett (U.S. equities, U.S. Treasuries, commodities, and cash each at 25%) has performed strongly this year, with an annualized return already reaching 16%, the best since 2021.

In today’s distorted setup where U.S. stock weighting is extremely concentrated in a handful of mega-caps, the portfolio’s outstanding performance powerfully demonstrates the unique value of diversified and dispersed investment.

Hartnett points out that, currently, the team is positioning strategically for bearish or bullish moves based on four major secular turning points from a long-cycle perspective. These four core investment themes each focus on commodities, emerging markets, small-cap stocks, and consumer stocks that will soon be formally incorporated into core allocations.

When assessing market risk appetite, Hartnett provides a clear line dividing bulls and bears.

As long as the “Big Tech seven sisters” ETF (MAGS) can hold steady above the 200-day moving average support (at $65), and the FX bellwether—AUD/JPY—remains above the 110 level, market capital will still prefer to buy on dips or rotate between sectors rather than withdraw risk assets in a big way.

However, a potential black swan is approaching. Current 30-year U.S. Treasury real yields have jumped to the highest level since November 2008, tightening the global financial conditions sharply. This high-rate environment will keep applying pressure until the Fed is ultimately forced to act—actively puncturing the asset bubble driven by AI speculation and suppressing inflation triggered by the wealth effect.

Risk disclosures and disclaimer

        The market involves risk; investment should be done cautiously. This article does not constitute personal investment advice, nor does it consider the special investment objectives, financial conditions, or needs of any individual user. Users should consider whether any opinions, viewpoints, or conclusions in this article fit their specific circumstances. Invest based on this at your own risk, and responsibility is borne by you.