The following is excerpted from a July 26 conversation between two partners. The discussion focuses on the outlook for semiconductors and AI, geopolitical developments in the Middle East and their market implications, the potential impact of the U.S. midterm elections, and Federal Reserve policy. It covers major asset classes including equities, fixed income, and gold.
Liu: Markets have experienced some volatility over the past two months. What have been the main drivers?
Fan: We flagged several risks in our previous video. Broadly speaking, three factors have driven both the recent volatility and current market conditions.
First, the market needed time to consolidate after its sharp rally, particularly in the semiconductor sector.
Second, geopolitical developments in the Middle East have increased uncertainty around the inflation outlook.
Third, against this macroeconomic backdrop, uncertainty surrounding the policy direction of the new Federal Reserve chair and the midterm elections has amplified market volatility.
Liu: Let’s examine these factors individually, beginning with semiconductors—arguably the most closely watched sector globally. Following the recent correction, do you expect prices to decline further or stabilize and recover?
Fan: As we noted two months ago, the semiconductor rally has been supported by the fundamentals of the AI industry. However, sentiment became overheated in the short term, and the market priced in a very strong future growth trajectory well in advance. A period of pronounced volatility or a meaningful correction should therefore come as no surprise, particularly given the cyclical nature of investor sentiment.
The sector’s trajectory following this correction will depend on two factors: the depth of the current drawdown and the evolution of industry supply and demand. One key consideration, for example, is whether data-center capital expenditure by major technology companies continues to grow more strongly than expected.
Liu: Is the principal risk that a slowdown in data-center spending by major technology companies could weaken semiconductor demand expectations and consequently weigh on share prices?
Fan: Yes. That is a risk that cannot be ignored, although it currently appears relatively unlikely.
Semiconductors are inherently cyclical. Rising demand attracts additional supply, which can eventually produce excess capacity. The cycle then repeats. This cycle, however, has been shaped by the infrastructure requirements of AI. As a result, price trends have exhibited a more structural growth pattern, with both the duration and magnitude of the upcycle materially exceeding those of a typical semiconductor cycle—hence the term “supercycle.”
Some analysts have compared the current semiconductor supercycle with that of the 1990s and concluded that the present cycle may be only halfway through. Under that scenario, the longer-term uptrend would resume following a period of correction and consolidation.
That said, such comparisons are inherently imperfect; history does not repeat itself mechanically. From a probabilistic perspective, there may be roughly a two-thirds chance that the market resumes its advance after the correction. But there is also perhaps a one-third chance that current valuations have already fully discounted future growth while underestimating the industry’s cyclical risks.
Liu: Given that uncertainty in the risk-reward profile, how should investors position their portfolios?
Fan: In my view, the more prudent approach is not to try to predict the semiconductor sector’s direction and make a concentrated, one-way bet. The period when the sector offered its most attractive asymmetry between potential return and downside risk has already passed. Any concentrated directional exposure therefore carries material risk.
A more balanced approach would be to hold broader exposure to the entire AI ecosystem—or simply to the U.S. technology sector. Investors could, for example, consider the Nasdaq-100 Index or the S&P 500 Information Technology sector.
Liu: So the underlying thesis is that structural growth across the broader AI ecosystem offers greater visibility, allowing investors to place less emphasis on risks within individual subsectors.
Fan: Exactly. The principal advantage is that the different components of the AI ecosystem can provide a natural internal hedge.
As we discussed, one semiconductor risk is that major technology companies may reduce capital expenditure. However, lower capital spending would strengthen those companies’ free cash flow and could support their share prices, helping offset weakness in the semiconductor segment.
As we have previously argued, the development of the AI industry—viewed as a whole—has been one of the principal drivers of the current bull market. That is unlikely to change over the next three to five years.
Liu: Let’s turn to the second source of uncertainty: Middle East geopolitics. We have discussed this subject in some detail before. Previously, we described the situation as one in which both sides were fighting while continuing to negotiate. That still appears to be the case, although tensions have escalated over the past two weeks. How do you expect the situation to develop, and what will be the implications for markets?
Fan: The fundamental logic surrounding the Strait of Hormuz remains unchanged: a prolonged closure would be detrimental to both the United States and Iran. At this stage, the dispute is primarily over nominal control.
Both sides need to support their respective political narratives while strengthening their negotiating positions. The latest round of U.S. strikes against Iran reflects the same logic. Both parties understand that military action is intended to create leverage at the negotiating table—not to force the other side into complete capitulation, which is not realistically achievable.
Our assessment of the broader trajectory is therefore unchanged. The situation is likely to stabilize gradually through a combination of intermittent conflict and negotiations. The most probable outcome remains one in which both sides declare victory and reach an implicit understanding on the key issues through third-party intermediaries, without signing an explicit agreement.
Transit through the Strait may remain intermittent during this process, but the eventual outcome should be a return to effective navigation.
Liu: How long could this combination of conflict and negotiation continue? Oil prices have recently rebounded. Can markets continue to absorb volatility in energy prices?
Fan: Developments in the Middle East affect markets principally through two channels: inflation and economic growth. So far, the economic impact of higher oil prices appears relatively limited, as we discussed in our previous video. The market’s main concern is the effect of oil prices on inflation, which is difficult to assess over the short term.
Oil futures, for example, surged at one point in April, but the inflation data did not show a corresponding increase. What we can do is monitor high-frequency inflation indicators. At present, these measures do not point to a clear acceleration in inflation.
In terms of timing, I believe pressure associated with the midterm elections will lead Trump to seek a new transit arrangement for the Strait of Hormuz in August. Iran is unlikely to reject such an initiative outright, and the parties should ultimately reach a practical understanding through intermediaries such as Oman.
As for oil prices, following the near-term rebound, they are more likely than not to resume a gradual decline.
The Federal Reserve is highly likely to leave policy unchanged at its July meeting. The September decision will depend on the inflation data available at that time. Markets are currently nervous and increasingly expect the Fed to raise rates in September. However, if tensions ease and oil prices decline in August, those expectations could shift. We believe that outcome is more likely.
Liu: Does that mean current market conditions could present an opportunity to buy on weakness?
Fan: At current levels, I do not believe the broader U.S. equity market has declined enough to justify aggressively buying the dip. The principal reason is that market leverage has risen materially over the past year, and historical evidence suggests that this excess may still need time to be absorbed.
However, if the market falls further because of short-term concerns over inflation, interest-rate increases, or similar uncertainties, that would indeed create a buying opportunity.
Investors should also be aware of seasonal patterns. Historical data suggest that markets generally experience increased volatility during the three to six months preceding a midterm election. From approximately two to three months before the election through the following six months, however, equities have historically tended to perform strongly. Based on this seasonal pattern, the probability of continued volatility between July and September is relatively high.
Liu: How do you assess the political uncertainty surrounding the midterm elections? Trump’s approval ratings have reportedly fallen to a historic low, largely because of public dissatisfaction with his decision to initiate a limited conflict with Iran. How significantly could this affect the midterms? Is there a risk that the Democrats win both chambers of Congress and reverse his existing policies?
Fan: From the perspective of the midterm elections, Trump’s decision to strike Iran was undoubtedly a major strategic error. He is now attempting to contain the political damage, but some of that damage cannot be reversed.
Current prediction-market data suggest that Republicans are almost certain to lose the House of Representatives. Their position in the Senate, however, remains considerably stronger. Although voters disapprove of Trump’s mistakes, they do not appear particularly satisfied with the available alternatives either.
For Democrats to mount a fundamental challenge to the Trump administration—for example, by removing Trump through the impeachment process—they would need to secure a substantial Senate majority. That currently appears highly unlikely.
The most probable outcome is that Democrats take control of the House while Republicans retain their Senate majority. A narrow Democratic Senate majority is also possible.
This is broadly the outcome Wall Street would prefer. Divided government would create stronger institutional checks and balances, reducing uncertainty around Trump’s policy agenda over the following two years.
Liu: Markets also tend to experience volatility during the first two or three months after a new Federal Reserve chair takes office. That timing happens to overlap with the midterm-election seasonality you mentioned.
Fan: That is correct. However, as we have emphasized repeatedly, the fundamental forces supporting the current bull market remain intact over the medium to long term.
These include continued growth in the AI industry, the combination of domestic reindustrialization and fiscal incentives, and a more stable Federal Reserve interest-rate policy. As long as these fundamentals remain unchanged, we will remain constructive on the equity-market outlook for the second half of the year, notwithstanding the potential for near-term volatility.
Liu: Our discussion has focused primarily on oil and equities. What is your outlook for other assets, such as gold, government bonds, and Bitcoin?
Fan: Our views are largely unchanged.
Gold is likely to remain range-bound and volatile. Based on historical patterns, any near-term rebound would more likely be followed by a continuation of the broader downtrend or sideways consolidation.
Long-term U.S. Treasury yields are likely to remain within a range of 4% to 5%. For China-based investors willing to hold these securities to maturity, they offer a relatively attractive yield.
Bitcoin and Ether have most likely entered a bottoming range. For long-term investors, current levels may represent an attractive entry point.