Zhongjin: In September, global assets face multiple challenges. It is recommended to increase allocations in Chinese and American tech stocks and gold when prices are low.

China International Capital Corporation (CICC) released a research report stating that global assets are facing multiple challenges in September. The issuance of AI-related bonds may accelerate, and the uncertainty surrounding U.S. Treasury rates is increasing; along with this, Waller has turned increasingly hawkish, raising the risk of a Federal Reserve interest rate hike; the U.S.-Iran conflict has intensified, leading to a significant rebound in oil prices. The bank advises maintaining confidence and patience, as the timing for policy adjustments may be approaching, and short-term fluctuations will not change the trend of global liquidity easing. Investors are encouraged to take advantage of market fluctuations to accumulate U.S.-China tech stocks and gold at lower prices. Key points from CICC are as follows: September may usher in a new round of AI financing impacts. In mid-August, long-term U.S. Treasuries underwent a rapid adjustment, with the 30-year Treasury yield briefly surpassing 5.3%. CICC previously indicated that the main reason behind this rise in long-term interest rates is not the increased supply of long-term Treasuries from the U.S. Treasury, but rather the expansion of AI-related credit bond supply. As tech giants continue to expand capital expenditures and lengthen financing durations, long-duration AI credit bonds compete for the same pool of long-duration funds as U.S. Treasuries, and the additional corporate bond supply raises the overall market term premium, leading to a passive increase in long-term Treasury yields. For the year as a whole, the bank expects net supply of investment-grade corporate bonds to approach $1 trillion by 2026, a year-on-year increase of over 70%, while net supply of interest-bearing Treasuries is expected to decline to $1.2 trillion, reflecting that the supply pressure in the bond market stems more from corporate bonds than from Treasuries. With the increased supply of AI-related bonds, the credit spreads of leading U.S. tech companies have also widened to varying degrees recently. September is the traditional issuance peak for U.S. investment-grade credit bonds, and the supply pressure on U.S. credit bonds may rise again. Bloombergs compilation of underwriters expects around $215 billion in investment-grade corporate bonds to be issued in September. In the context where long-term funding has already been largely consumed, a new round of concentrated issuance may once again raise financing costs and term premiums for credit bonds, causing disruptions to long-term U.S. Treasuries. Meanwhile, the stock market is also facing a degree of financing pressure. This year, the cumulative fundraising from U.S. IPOs has reached about $137.6 billion, a year-on-year increase of 464%, with SpaceX raising about $85.7 billion in a single offering, becoming the largest IPO in U.S. history. Equity financing related to AI is still heating up this autumn; Anthropic submitted its IPO application in June and is currently preparing for its listing, with the market expecting it to launch its IPO as early as October. If there is a concentrated release of AI financing, bond financing may increase long-duration supply and disrupt long-term rates, while IPOs could divert risk capital, placing pressure on highly valued assets. Increasing geopolitical risks and rising oil prices may delay the improvement of inflation, but core inflation may remain low. Since late August, the situation in the Middle East has escalated again, with the volume of commodity shipping through the Strait of Hormuz dropping to its lowest level since May; conflicts have resumed between Saudi Arabia and the Houthis, leading to attacks on Saudi energy facilities and raising market concerns about energy supply shortages. As a result, Brent crude has continued to rise from a stage low of $87.8 per barrel on August 26, breaking the $100 per barrel mark in intraday trading on September 9. CICC believes that rising oil prices may drive a temporary rebound in the nominal CPI in August, but core inflation is still expected to continue cooling. The U.S. CPI for August is set to be announced on September 11 (Friday), and CICC's macro asset allocation team predicts that nominal CPI in August may rise month-on-month to 0.36% (previously 0.07%, consensus estimate 0.4%), while year-on-year it remains at 3.37%; core CPI is expected to maintain a low month-on-month increase of 0.18% (previously 0.22%, consensus estimate 0.2%) and decrease year-on-year to 2.35%. The rebound in nominal inflation primarily stems from energy items: August is typically a month of seasonal decline in oil prices, but this year saw an unexpected seasonal rise in oil prices, pushing up the nominal CPI. Core CPI may remain low, mainly due to two factors: on one hand, high-frequency data shows that wholesale prices of used cars have deepened their declines over the past two months, which transmits to retail prices and may slow the month-on-month growth rate of used car CPI; on the other hand, falling import prices and tariffs are also putting downward pressure on inflation for other core goods. The market generally views Fridays CPI release as the decisive data for whether the Federal Reserve will raise rates in September, but if the above predictions hold true, the CPI may not provide a clear signal before the Federal Reserves meeting. If Waller wishes to raise rates, he can emphasize the month-on-month rebound in nominal inflation while noting that year-on-year figures have stopped improving. According to the logic presented at the Jackson Hole conference, as long as inflation is declining "not fast enough," a rate hike is warranted. If Waller does not wish to raise rates, he can highlight that core inflation is only 0.2% month-on-month and is still on a slight downward trend year-on-year. The same data can lead to entirely different policy implications depending on the perspective taken. Of course, considering statistical error, the bank acknowledges that their predictions may also be incorrect. If inflation is significantly above or below the aforementioned projections, this could provide a clearer policy signal before the September Federal Reserve meeting. Federal Reserve Chair Waller has clearly adopted a hawkish stance at Jackson Hole, increasing the risk of a rate hike in September. Waller previously argued that if inflation remains persistently high, further rate hikes may be necessary, and has now further stated that inflation must not only decrease but must do so clearly and at a sufficiently rapid pace toward the 2% target; otherwise, the Federal Reserve must take action. This marks the first time Waller has set a clear requirement for the pace of inflation reduction, significantly lowering the bar for rate hikes. Unlike the rapid decline in inflation seen over the past two months, the recent rebound in oil prices is likely to slow the pace of inflation reduction, heightening the risk of a rate hike in September. Another piece of incremental information from the Jackson Hole meeting is that Waller refuses to view the recent improvement in data as a shift in trend. Recently, the U.S. labor market has cooled somewhat from earlier periods, and consumption growth is exhibiting signs of marginal slowing; CPI has cooled for two consecutive months, but Waller still emphasizes that the potential for inflation improvement is limited, with labor market cooling primarily stemming from a contraction in labor supply, while the economy and job market remain resilient. Waller's hawkish interpretation of dovish data may be aimed at restoring the credibility of the dollar. The bank believes that the fundamentals of the U.S. economy do not actually support a rate hike: the potential inflation anchor in the U.S. is not high, the increase in oil prices has not created a significant second-round effect, and inflation is still expected to continue returning towards 2%. The job market is also cooling, and with mid-term elections approaching, any further tightening faces political constraints. Therefore, Waller may merely be using hawkish rhetoric this time, and not raising rates in September remains the bank's baseline scenario, but the possibility of the Federal Reserve taking action to overcorrect to restore credibility cannot be ruled out: due to the clearly lowered bar for rate hikes articulated at the Jackson Hole meeting, the Federal Reserve may be faced with a dilemma in the September meeting; not raising rates would undermine policy credibility, while raising rates could further damage the economy and increase political costs, contributing to high policy uncertainty. Markets have already priced in rate hikes from European and Japanese central banks in September; if the Federal Reserve also embarks on rate hikes, the Fed, the European Central Bank, and the Bank of Japan would tighten synchronously, creating a phase-shifting shock to the global liquidity environment. The timing for policy adjustments may be gradually approaching, and the pressure for AI bond issuance may phase out after September. The bank believes that multiple policy missteps are occurring simultaneously in the U.S. If economic and market conditions continue to deteriorate, policy adjustments may occur at any time: First, with only two months left until the mid-term elections, the optimal solution for the U.S. is to quickly end conflicts and alleviate oil price and inflation pressures. Second, over the past two months, the U.S. has seen slowdowns in employment, consumption, and inflation. Any rebound in employment and inflation data in September is unlikely to be sustainable; U.S. inflation is highly likely to drop to around 2% next year, and the AI revolution will lower the long-term inflation anchor; in fact, the Federal Reserve has ample grounds to wait for inflationary pressures to ease naturally, allowing them to hold off on rate hikes and implement rate cuts after inflation drops. Even if a rate hike occurs in September, there may still be a quick correction in the future, and the path of monetary policy may return to rate cuts. Whether or not there is a rate hike in September, future monetary policy may still lean toward easing. Third, the pressure from AI bond financing may marginally ease after September. September is traditionally a peak issuance month for U.S. credit bonds, and the bank anticipates that entering October and November, issuance pressure will likely decrease seasonally, potentially reducing disruptions to term premiums and long-term rates. Investors should accumulate U.S.-China tech stocks and gold at lower prices, with U.S. Treasuries being relatively lower on the priority list; the September FOMC meeting is a key moment. In summary, the recent pullback may create opportunities for increased allocations, with the key being when a turning point will emerge. If the FOMC meeting on September 16 results in a rate hike, there may not be further hikes in the future, and the market could begin pricing in a complete exhaustion of negative impacts; stocks and gold may initially drop and then rise after the meeting. If there is no rate hike in September, the immediate risk of one may abate, and the Federal Reserve's previously aggressive hawkish stance without action may weaken policy credibility, ultimately benefiting gold. From an event timing perspective, there may be a high risk-reward rebound window following the September FOMC meeting, and it is recommended to focus on this. At the same time, considering the uncertainties of the U.S.-Iran situation and economic data, the market may also start a rebound before the FOMC meeting (for example, if Trump quickly resolves the conflict, or if U.S. CPI falls significantly short of expectations, or if other significant policy adjustments occur), hence trading strategies should maintain a degree of flexibility. Given that the bullish trends for gold and U.S.-China tech stocks have not changed, the bank believes there is no need to mechanically wait for a specific time point; if the market shows a clear pullback in the coming weeks, gradual accumulation of stocks and gold may begin even before the Federal Reserve meeting. In terms of asset categories: (1) Gold remains the clearest direction for increased allocation. Whether the Federal Reserve ultimately shifts to easing to improve liquidity, or recent policy missteps harm the credibility of the dollar, the mid-term logic for gold remains unchanged, and pullbacks actually present opportunities for accumulation. (2) U.S.-China stocks can also be added to at lower prices, particularly in the tech sector. The trend of the AI industry has not reversed, and recent adjustments have stemmed more from pressures in funding and sentiment rather than a fundamental deterioration of profitability; if policy and liquidity pressures ease, overvalued assets are likely to have greater rebound elasticity. (3) There is no rush to bottom-fish U.S. Treasuries. The risks of AI financing and potential rate hikes in September may continue to disrupt long-term rates, with the certainty of long-term bonds being lower than that of gold and stocks. Compared to betting on a sharp decline in long-term rates, the bank's stance on U.S. Treasuries leans towards neutrality, suggesting patience in waiting for the release of policy risks and supply pressures.