CICC: Global assets faced multiple challenges in September, it is recommended to increase the allocation of Chinese and US technology stocks and gold at dips
The Zhitong Finance App learned that CICC released a research report saying that global assets faced multiple challenges in September. AI debt issuance is likely to accelerate, and interest rate variables on US bonds will increase; the risk of the Fed's interest rate hike will rise; the US-Iran conflict will intensify, and oil prices will clearly rebound. The bank suggests maintaining confidence and patience. The time for policy correction may be getting closer. Short-term fluctuations will not reverse the trend of global liquidity easing, and use market fluctuations to increase the allocation of Chinese and US technology stocks and gold at dips.
CICC's main views are as follows:
September may welcome a new round of AI financing shocks
In mid-August, long-term US bonds experienced a round of rapid adjustments, and the yield on 30-year US bonds once exceeded 5.3%. The bank previously pointed out that the main reason behind this round of rising long-term interest rates is not an increase in the supply of long-term treasury bonds by the US Treasury, but rather an expansion in the supply of AI-related credit bonds. As technology giants' capital spending continues to expand and financing periods lengthen, long-term AI credit bonds compete with US bonds for the same batch of long-term capital, and the supply of additional corporate bonds raises the maturity premium of the entire market, driving a passive upward trend in interest rates on long-term US bonds.
Looking at the full year, the bank expects the net supply of investment-grade corporate bonds to be close to 1 trillion US dollars in 2026, up more than 70% year on year. Instead, the net supply of interest-paying treasury bonds will fall to 1.2 trillion US dollars, reflecting the supply pressure on the bond market not on US bonds, but on corporate bonds.
As the supply of AI-related bonds has increased, credit spreads for leading US tech companies have also widened to varying degrees recently.
September is the traditional peak season for US investment-grade credit bonds, and the pressure on the supply of US credit bonds may rise again. Underwriters summarized by Bloomberg are expected to issue approximately $215 billion in investment-grade corporate bonds in September.
In a context where long-term long-term capital has already been consumed in large quantities, a new round of centralized issuance may once again raise the financing costs and term premiums for credit bonds, causing disturbances for long-term US bonds. At the same time, the stock market is also facing some financing pressure. Since this year, US IPOs have raised a total of about 137.6 billion US dollars, an increase of 464% over the previous year. Of these, SpaceX raised about 85.7 billion US dollars in a single transaction, making it the largest IPO in US history.
AI-related equity financing is still heating up in the fall. Anthropic submitted an IPO application in June and is currently preparing for listing. The market expects it to launch an IPO in October as soon as possible. If AI financing is released centrally, bond financing may increase long-term supply and disrupt long-term interest rates, while IPOs may divert risk capital. Together, the two will put pressure on overvalued assets.
Geographic risks are rising, and rising oil prices are slowing down the progress of improving inflation, but core inflation may remain low
Since the end of August, the situation in the Middle East has once again escalated, and commercial ship traffic in the Strait of Hormuz fell to its lowest level since May; clashes broke out again between Saudi Arabia and the Houthis, causing Saudi energy facilities to be attacked, raising market concerns about the shortage of energy supply. Affected by this, Brent crude oil continued to rise from a phase low of 87.8 US dollars/barrel on August 26, and broke through 100 US dollars/barrel in the intraday period on September 9.
The bank believes that the rise in oil prices may drive a phased rebound in US nominal CPI in August, but core inflation is still expected to cool down. On September 11 (Friday), the US CPI for August will be announced. The CICC Asset Allocation Team predicts that in August, the US nominal CPI will rebound to 0.36% month-on-month (previous value 0.07%, consistent forecast 0.4%), and maintain a year-on-year low of 3.37%; the core CPI may maintain a low 0.18% month-on-month (previous value 0.22%, consistent forecast 0.2%), and drop to 2.35% year on year.
The rebound in nominal inflation mainly comes from the energy sector: August is usually a month of seasonal decline in oil prices, but this year oil prices rose against the season, driving up nominal CPI.
The core CPI is likely to remain low, mainly due to two factors: on the one hand, high-frequency data shows that the decline in used car wholesale prices has deepened in the past two months, which has spread to retail prices, or caused the month-on-month growth rate of used car CPI to slow down; on the other hand, falling import prices and tariffs are also depressing inflation for other core commodities.
The market generally sees Friday's CPI as decisive data on whether the Fed will raise interest rates in September, but if the above forecast is fulfilled, the CPI may not provide a clear signal before the Federal Reserve meeting. If Walsh wants to raise interest rates, he can emphasize that nominal inflation has rebounded month-on-month and has stopped improving year over year. According to the logic of Jackson Hole's speech at the annual conference, as long as inflation falls “not fast enough,” interest rates should be raised. If Walsh doesn't want to raise interest rates, he can emphasize that core inflation is only 0.2% month-on-month, and is still declining slightly year over year.
The same data, different perspectives, can draw diametrically opposite policy implications. Of course, considering the statistical error, the line's predictions are also likely to be wrong. If inflation is significantly higher or lower than the above forecast, a more clear policy signal may be given before the September Federal Reserve meeting.
Federal Reserve Chairman Walsh clearly turned Hawk in Jackson Hole. The risk of the Fed's interest rate hike in September is rising
Walsh previously argued that “if inflation continues to be high, further interest rate hikes may be needed,” but this time he further proposed that inflation must not only fall, but also fall “clearly and fast enough” to the 2% target; otherwise, the Federal Reserve still needs to take action. For the first time, Walsh put forward clear requirements for the rate of decline in inflation, and the threshold for raising interest rates was lowered markedly.
Unlike the rapid decline in inflation in the past 2 months, due to the recent rebound in oil prices, the rate of decline in inflation would have slowed down, increasing the risk of interest rate hikes in September. Another incremental message from the Jackson Hole conference was that Walsh declined to view recent data improvements as a trend change. Recently, the US labor market has cooled down compared to the previous period, and consumption growth has also shown signs of a marginal slowdown. CPI has cooled for two consecutive months, but Walsh still emphasized that potential improvements in inflation are limited. The slowdown in employment is mainly due to a contraction in labor supply, and the economy and job market are still resilient.
Walsh forcibly interprets dovish data from a hawkish perspective, probably to repair the dollar's credibility. The bank believes that the fundamentals of the US economy actually do not support interest rate hikes: the potential center of US inflation is not high, and the rise in oil prices has not had a clear “secondary effect”. In the future, inflation is expected to continue to fall to 2%, and the job market is also cooling down. As the midterm elections approach, further austerity is also facing political constraints.
Therefore, Walsh may just be rhetoric this time. Not raising interest rates in September is still the bank's benchmark scenario, but now it is impossible to rule out the possibility that the Fed will “overdo” interest rate hike in order to restore its credibility: since the threshold for interest rate hikes was significantly lowered at the Jackson Hole meeting, the September Fed meeting may face a dilemma. Not raising interest rates harms policy credibility. Raising interest rates will further hurt the economy and increase political costs, and policy uncertainty is high.
Previously, the market had already taken into account the September rate hikes in Europe and Japan. If the Federal Reserve also starts raising interest rates, the Federal Reserve, the European Central Bank, and the Bank of Japan will simultaneously tighten, causing a phased impact on the global liquidity environment
The time for policy correction may gradually approach, and the pressure on AI debt issuance may ease in stages after September
The bank believes that many current US policies have gone wrong at the same time. If the economic and market situation continues to deteriorate, the policies may be corrected at any time:
First, there are currently only 2 months left until the midterm elections. America's best solution is to end the conflict as soon as possible and reduce the pressure on oil prices and inflation.
Second, US employment, consumption, and inflation have all been slowing down in the past two months. The rebound in some employment and inflation data in September is unsustainable. US inflation is likely to fall to around 2% next year, and the AI revolution will depress the center of long-term inflation. In fact, the Fed is fully in a position to wait for inflationary pressure to ease naturally. It can choose not to raise interest rates now and start cutting interest rates; even if interest rates are raised in September, the future may quickly correct and restart interest rate cuts. Regardless of whether interest rates are raised in September or not, monetary policy may return to cutting interest rates in the future.
Third, the pressure on AI bond financing may ease marginally after September. September is the traditional peak season for US credit bonds. The bank expects that after entering October-November, issuance pressure is expected to decline seasonally, and disturbances in term premiums and long-term interest rates may ease.
The allocation of Chinese and US technology stocks and gold was increased on dips. The ranking of US bonds was relatively low. The September FOMC meeting was a critical point
Based on the above analysis, the recent pullback may create opportunities to increase allocations at dips. The key is when the turning point will occur.
If interest rates are raised at the FOMC meeting on September 16, interest rates may not continue to be raised in the future, the market may run out of profit, and stocks and gold may fall and then rise after the meeting. If interest rates are not raised in September, the risk of short-term interest rate hikes will be directly mitigated, while the Federal Reserve has taken no action. On the contrary, it may further weaken the credibility of the policy, which is more beneficial to gold.
Judging from the pace of events, after the FOMC in September, there may be a rebound window with relatively high risk and return, so it is recommended to focus on it. At the same time, considering the situation between the US and Iran and the uncertainty of economic data, the market may also start a rebound before the FOMC meeting (such as Trump quickly ending the conflict, or the US CPI falling far short of expectations, or other major policy adjustments), and the pace of transactions needs to be kept flexible. Since the bullish trend of gold and Chinese and US technology stocks has not changed, the bank believes that there is no need to wait mechanically for a specific point in time. If the market clearly recovers in the next few weeks, it can also consider starting to gradually increase the allocation of stocks and gold before the Federal Reserve meeting. By asset class:
(1) Gold is still the clearest direction of overmatching. Whether the Federal Reserve finally switched to easing to improve liquidity, or recent policy mistakes damage the dollar's credibility, the medium-term logic of gold has not changed; the pullback instead provides an opportunity to increase allocations.
(2) The allocations of Chinese and US stocks can still increase due to low allocations, especially in the technology sector. Trends in the AI industry have yet to be reversed. Recent adjustments are more due to financial and emotional pressure rather than a fundamental deterioration in profit trends; if policy and liquidity pressures ease, overvalued assets will have greater resilience.
(3) US debt is in no hurry to bottom out. The risk of AI financing and interest rate hikes in September may continue to disrupt long-term interest rates, and the certainty of long-term bonds is lower than that of gold and stocks. Compared to betting on a rapid decline in long-term interest rates, the bank's views on US bonds are more neutral, and it is recommended to wait patiently for policy risks and supply pressure to be released.