Tom Lee's latest interview: The bull market is far from over, and Ethereum will rise to $12,000 next year.

CN
5 hours ago

Original Title: Tom Lee: The Biggest Market Shake-Up Is Coming

Original Author: Fundstrat

Original Translation: Yuliya, PANews

In an era where most analysts hold a cautious or even pessimistic view of the market, Tom Lee, Chairman of BitMine and a senior strategist, has issued a distinctly bullish tone. In this interview with Fundstrat, Tom Lee delves into the current macro cycle, the AI super cycle, changes in market sentiment, inflation risks, and the future trajectory of crypto assets. Tom Lee believes that the market is at a critical juncture of a "super cycle," and that investors' misjudgments regarding macro signals, the yield curve, inflation logic, and the AI industry cycle are leading to systemic mismatches. He not only predicts that the U.S. stock market will reach 7000 to 7500 points by the end of the year but also points out that Ethereum and Bitcoin are expected to see a strong rebound. PANews has compiled and translated this dialogue.

We Are in a Misunderstood "Super Cycle"

Host: Tom, welcome. Let's first review that the market has risen by more than 80% over the past three years. You have consistently been one of the few bullish voices. In your view, what have the 90% of analysts and bears gotten wrong from 2023 to 2024, and even this year?

Tom Lee: 80% of trading essentially depends on the macro environment. Over the past three years, investors have largely considered themselves "macro traders," but they have made two key mistakes.

  • First, they have been overly reliant on the "scientific nature" of the yield curve. When the yield curve inverts, everyone assumes it signals a recession. However, we at Fundstrat have explained that this inversion is due to inflation expectations—short-term inflation is high, so short-term nominal interest rates should be higher, but in the long term, they will decline, which is the reason for the curve inversion.

  • Second, our generation has never truly experienced inflation, so everyone uses the "stagflation" of the 1970s as a template without realizing that we do not have the tricky conditions today that would cause such persistent inflation.

As a result, people are structurally bearish, believing that "an inverted curve means recession, and stagflation is imminent." They completely miss that companies are dynamically adjusting their business models in real-time to cope with inflation and the Fed's tightening policies, ultimately delivering excellent earnings. In the stock market, time is the best friend of great companies and the worst enemy of mediocre ones, whether during inflationary periods or bull markets; this is a truth.

Host: I notice that you believe the current market environment is somewhat similar to 2022, when almost everyone turned bearish, and now that anxiety is resurfacing in the market, you are once again taking a bullish stance. What do you think is the biggest misunderstanding people have about the current market landscape?

Tom Lee: I think the hardest thing for people to understand and grasp is the "super cycle." We turned structurally bullish in 2009 because our cycle research indicated that a long-term bull market was beginning. In 2018, we identified two future super cycles:

· Millennials: They are entering their prime working age, which will provide a strong tailwind for the next 20 years.

· Global labor shortage in the golden age: This may sound mundane, but it is this factor that lays the foundation for the boom in artificial intelligence (AI).

Why the AI Boom is Essentially Different from the Internet Bubble?

Tom Lee: We are in a boom driven by AI, which is causing asset prices to rise continuously. This is actually a textbook case: from 1991 to 1999, there was a labor shortage, and tech stocks boomed; from 1948 to 1967, there was also a labor shortage, and tech stocks thrived. Today, the AI wave is replaying this pattern.

But the problem is that many people view high Sharpe ratio stocks as a bubble and try to short companies like Nvidia, but this perspective may be flawed. They forget that today's AI industry is fundamentally different from the internet of the 1990s. Back then, the internet was merely a "capital expenditure frenzy," while AI represents a "gain of function."

Host: Many people compare the current AI boom to the internet bubble of the late 1990s, even likening Nvidia to Cisco back then. You experienced that era firsthand; what do you think are the essential differences between the two?

Tom Lee: This comparison is interesting but fundamentally flawed. The lifecycle of telecom capital expenditure (Cisco) and GPUs (Nvidia) is completely different.

People forget that in the late 1990s, the core of the capital expenditure boom was the telecom industry—laying fiber optics, not the internet itself. At that time, telecom spending in emerging markets was tied to GDP growth, and this wave spread to the U.S., leading companies like Quest to lay fiber optics madly along railways and under streets, while Global Crossing laid submarine cables globally. The problem was that the internet's consumption of these fibers could not keep pace with their installation speed; at the peak, nearly 99% of the fiber was idle "dark fiber."

Today, however, the situation is completely the opposite. The market demand for Nvidia chips remains strong, with their GPUs currently operating at nearly 100% utilization, ready to use and fully meeting market demand. The supply of Nvidia chips cannot meet demand; even if production capacity increases by 50%, all chips will still be sold out quickly. The industry currently faces three major constraints: Nvidia chip supply, related silicon materials, and energy supply. These factors collectively restrict the market's expansion speed. Meanwhile, the pace of functional improvements in AI technology far exceeds expectations, further intensifying the demand for hardware. However, capital expenditure has yet to keep pace with this trend, and the industry as a whole remains in a state of supply shortage. In other words, AI capital expenditure is still "lagging behind the pace of innovation."

Year-End Market Predictions and the Potential of Cryptocurrencies

Host: You have mentioned several times that the S&P 500 index could reach 7000 or even 7500 points by the end of the year. In your optimistic year-end outlook, which sector do you think will bring the biggest surprises?

Tom Lee: First, market sentiment has become quite pessimistic over the past few weeks. The government shutdown temporarily withdrew funds from the economy, and the Treasury has not disbursed payments, leading to a liquidity contraction and market volatility. Whenever the S&P 500 index drops by 2-3%, or AI stocks drop by 5%, people become very cautious. I believe the foundation of bullish sentiment is very unstable; everyone feels that the market top is approaching. But I want to emphasize: when everyone thinks the top is near, it cannot form. The top of the internet bubble formed because no one believed stocks would fall.

Secondly, you must remember that the market has performed strongly over the past six months, but people's positions have deviated significantly, indicating a huge potential demand for stocks. In April of this year, due to tariff issues, many economists declared that a recession was imminent, and institutional investors traded accordingly, effectively preparing for a massive bear market. This mispositioning cannot be adjusted in just six months.

Now that we are approaching year-end, 80% of institutional fund managers are underperforming their benchmarks, marking the worst performance in 30 years. They have only 10 weeks left to catch up, which means they will have to buy stocks.

So, I believe several things will happen before the end of the year:

· AI trading will make a strong comeback: Despite recent fluctuations, the long-term outlook for AI has not been affected, and companies are expected to make significant announcements as they look towards 2026.

· Financial stocks and small-cap stocks: If the Fed cuts rates in December, confirming its entry into a loosening cycle, this will be very beneficial for financial and small-cap stocks.

· Cryptocurrencies: Cryptocurrencies are highly correlated with tech stocks, financial stocks, and small-cap stocks. Therefore, I believe we will also see a large-scale rebound in cryptocurrencies.

Host: Since you mentioned cryptocurrencies, what level do you think Bitcoin will reach by the end of the year?**

Tom Lee: Expectations for Bitcoin have lowered; part of the reason is that it has been consolidating, and some early Bitcoin holders (OGs) sold when the price exceeded $100,000. But it remains a severely underweighted asset class. I believe Bitcoin has the potential to reach the high tens of thousands by the end of the year, and it could even reach $200,000.

However, to me, it is more evident that Ethereum may see significant gains before the end of the year. Even "Cathie Wood" has written that stablecoins and tokenized gold are eating into Bitcoin's demand. Both stablecoins and tokenized gold operate on smart contract blockchains like Ethereum. Additionally, Wall Street is actively positioning itself, with BlackRock's CEO Larry Fink hoping to tokenize everything on the blockchain. This means that expectations for Ethereum's growth are rising. Our head of technical strategy, Mark Newton, believes that by January next year, Ethereum's price could reach $9,000 to $12,000. I think this prediction is reasonable, which means Ethereum's price will more than double from now until the end of the year or January next year.

Overestimated Inflation and Controllable Geopolitics

Host: You mentioned that the Fear and Greed Index closed at 21 last Friday, in the "extreme fear" range; the CME FedWatch Tool shows a 70% probability of a rate cut in December. Do you think this performance pressure will also prompt institutional funds to flow into cryptocurrencies like Ethereum and Bitcoin?

Tom Lee: Yes, I believe it will. Over the past three years, the S&P 500 has recorded double-digit gains for three consecutive years, and this year it may even exceed 20%. At the end of 2022, almost no one was bullish. At that time, wealthy individuals and hedge funds were advising clients to turn to cash or alternative assets—private equity, private credit, venture capital—only to see these asset classes underperform compared to the S&P. This "mismatch" is backfiring on institutions.

Therefore, 2026 should not be viewed as a bear market year. On the contrary, investors will chase high-growth stocks like Nvidia again, as their earnings are still growing over 50%.

Meanwhile, the crypto market will also benefit. Although the market generally believes that Bitcoin's four-year cycle is about to end and should enter a correction, this judgment overlooks the macro environment. But they forget that the Fed is about to start cutting rates. Our research shows that the correlation between the ISM Manufacturing Index and Bitcoin prices is even higher than that of monetary policy. It is difficult for Bitcoin to peak before the ISM index reaches 60.

Currently, the cryptocurrency market is constrained by insufficient monetary liquidity, and the Federal Reserve's quantitative tightening (QT) is expected to end in December, but there has been no clear signal of easing, which has left investors confused. However, as these macro factors gradually become clearer, the cryptocurrency market is expected to see more positive performance.

Host: What do you think is the most overestimated risk in the current market?

Tom Lee: I believe the most overestimated risk is the "return of inflation." Too many people think that monetary easing or GDP growth will create inflation, but inflation is a very mysterious thing. We have experienced years of loose monetary policy without inflation. Now, the labor market is cooling, and the housing market is weakening; the three main drivers of inflation—housing, labor costs, and commodities—are not rising. I even heard a Federal Reserve official say that core services inflation is rising, but after we checked, we found that this was completely wrong. The PCE core services inflation is currently running at 3.2%, below the long-term average of 3.6%. Therefore, the view that inflation is strengthening is incorrect.

Host: If an unexpected situation arises, such as geopolitical tensions, war, or supply chain issues causing oil prices to soar, would that be a variable that would make you turn bearish?

Tom Lee: There is indeed that possibility. If oil prices rise high enough to cause a shock. Looking back at the last three economic shocks not caused by the Federal Reserve, they were all commodity price shocks. However, for oil to become a heavy burden on households, its price needs to reach a very high level. Over the past few years, the energy intensity of the economy has actually decreased.

So, oil prices need to approach $200 to cause such a shock. We have previously approached $100 oil prices, but it did not cause a shock. You really need oil prices to triple. This summer, when the U.S. bombed Iran's nuclear facilities, some predicted that this would lead to oil prices soaring to $200, but in reality, oil prices hardly fluctuated.

Host: Yes, geopolitics has never long-term weighed down the U.S. economy or the U.S. stock market. We have had localized shocks, but the U.S. has never experienced a real economic recession or a massive stock market crash due to geopolitics.

Tom Lee: Absolutely correct. Geopolitics can destroy unstable economies. But in the U.S., the key question is: Will corporate earnings collapse due to geopolitical tensions? If not, then we should not use geopolitics as a primary reason to predict a bear market.

Overcoming Fear and Greed

Host: If Federal Reserve Chairman Powell unexpectedly does not cut rates in December, how would the market react?

Tom Lee: In the short term, that would be negative news. However, while Chairman Powell is doing a good job, he is not popular in the current administration. If he does not cut rates in December, the White House may accelerate plans to replace the Federal Reserve Chairman. Once replaced, the Fed may see the emergence of a "shadow Fed," and this new "shadow Fed" will establish its own monetary policy. Therefore, I believe the negative impact will not be so lasting, as the new chairman may not have to be constrained by various voices within the Fed, and the execution of monetary policy may change.

Host: I have many friends who have been holding cash since 2022 and are now very conflicted, fearing the market is too high while also fearing missing out on more gains. What is your advice for this dilemma?

Tom Lee: This is a great question because many people face this dilemma. When investors sell stocks, they actually need to make two decisions: first, to sell, and second, when to re-enter the market at a better price. If they cannot ensure they can tactically re-enter the market, panic selling may lead to missing out on long-term compounding returns. Investors should avoid panic selling stocks due to market fluctuations; every market crisis is actually an investment opportunity, not a selling opportunity.

Secondly, for investors who have already missed market opportunities, I recommend gradually re-entering the market through a "dollar-cost averaging" approach rather than investing a lump sum. It is advisable to spread investments over 12 months or longer, investing a fixed percentage each month, so even if the market declines, they can achieve a better cost advantage through phased buying. They should not wait to enter the market after a market adjustment, as many investors hold similar thoughts, which may lead to further missed opportunities.

Host: How do you view the roles of retail and institutional investors? Some believe this bull market is primarily driven by retail investors.

Tom Lee: I want to correct a common misconception: retail investors do not perform worse than institutional investors in the market, especially those individuals with a long-term investment perspective. Many retail investors invest in stocks based on a long-term view, which makes it easier for them to accurately grasp market trends. In contrast, institutional investors, due to the need to outperform peers in the short term, often tend to time the market, potentially overlooking the long-term value of certain stocks. Anyone operating in the market with a long-term perspective can be considered "smart money," and this type of investor is more concentrated among retail investors.

Host: For companies like Palantir, which have price-to-earnings ratios in the triple digits, many believe they are too expensive. Under what circumstances do you think a triple-digit P/E ratio is still reasonable for long-term investors?

Tom Lee: I categorize companies into two circles: the first circle consists of those that are not profitable but have a P/E ratio of 100 (about 40% of the 4,000 publicly listed companies outside the main market), and most of these are poor investments.

The second circle is what we call N=1 companies:

  1. These companies are either laying the groundwork for a huge long-term story, so they are not profitable now;

  2. Or their founders are continuously creating new markets, making current profit flows unable to reflect their future.

Tesla and Palantir are examples. They should enjoy extremely high valuation multiples because you are discounting their future. If you insist on only paying Tesla a P/E ratio of 10, you will miss out on opportunities over the past seven or eight years. You need to think differently to find these unique, founder-driven companies.

Lessons Learned and Final Advice

Host: Many people say that this rebound is too concentrated in a few stocks, like Nvidia, which is a huge sign of a bubble. Do you agree with this view?

Tom Lee: Artificial intelligence is a scalable business, which means you need to invest huge amounts of money. You and I cannot create a product in a garage that competes with OpenAI.

Scalable industries are like energy or banking. There are only eight major oil companies in the world. If someone says oil is a cyclical business because there are only eight companies buying oil, we would find that absurd. Because you must be big enough to drill for oil. AI is the same; it is a scalable business. This is what the current market landscape shows. Do we really want Nvidia to deal with thousands of small companies? I would prefer them to collaborate with large companies that can deliver results and ensure financial viability. So, I think the current concentration phenomenon is logical.

Host: Despite having worked in this industry for forty years, what is the most important lesson the market has taught you personally over the past two years?

Tom Lee: The past two years have shown that the public's "collective misreading/misunderstanding" can last a long time, as we discussed at the beginning, many people firmly believe in a recession due to the inverted yield curve, even though corporate data does not support it; they prefer to believe their anchoring beliefs. Companies have thus become cautious and adjusted their strategies, but earnings remain excellent. Many times, when data conflicts with their views, people choose to believe themselves rather than the data.

The reason Fundstrat has maintained a bullish outlook is that we do not cling to our views; we anchor on earnings, and earnings data ultimately proves everything. People call us "permanent bulls," but earnings themselves have been "permanently rising," what more can I say? We are simply following a different set of data that ultimately drives stock prices.

Distinguish between "conviction" and "stubbornness." Stubbornness is thinking you are smarter than the market; conviction is being steadfast based on the right things. Remember, in front of a room full of geniuses, you can only be average at best.

Host: Peter Lynch said, "Waiting to correct loss money is more than correcting the loss itself." What do you think?

Tom Lee: There are a few masters of contrarian trading in the market, such as Peter Lynch, David Tepper, and Stan Druckenmiller, who are good at making decisive decisions when market sentiment is low. Take Nvidia as an example; when its stock price fell to $8, many were too afraid to buy, and then every 10% drop further exacerbated investors' hesitation. Tom Lee believes that this emotional stubbornness often stems from a lack of firm conviction rather than rational judgment.

Host: How do you explain the emotional, rather than fundamental, reactions exhibited by many investors when the market declines?

Tom Lee: This is a behavioral issue. "Crisis" is composed of "danger" and "opportunity." Most people only focus on danger during a crisis. When the market declines, people only think about the risks facing their investment portfolios or believe, "Oh my, I must have missed something because the good idea I firmly believed in should be going up every day."

But in reality, they should see this as an opportunity because the market will always give you chances. The period from February to April this year during the tariff crisis is a good example. Many people went to the other extreme, believing we were heading for a recession or that everything was over, but they only saw the danger and not the opportunity.

Additionally, the influence of emotions and political biases on market views is significant. Earlier consumer sentiment surveys showed that 66% of respondents leaned towards the Democratic Party, and these respondents reacted more negatively to the economy, while the stock market could not recognize this political affiliation difference. Companies and the market itself are independent of political views; investors need to transcend emotions and political biases. The "fan mentality" and self-esteem factors in investing can lead to decision-making biases, such as investors tending to bet on their favorite companies or feeling validated when stocks rise, while feeling frustrated when they fall. Even machines cannot completely eliminate biases, as their designs still carry human characteristics. To better cope with these influences, investors should focus on super cycles and long-term trends, such as Nvidia or Palantir's missions in the AI field, as short-term price fluctuations do not change their long-term potential.

Host: Finally, if you had to describe the stock market for the next 12 months in one sentence, what would you say?

Tom Lee: I would say, "Buckle up."

Because over the past six years, although the market has risen a lot, we have experienced four bear markets. This means we will almost experience a bear market every year, which will test your resolve. So I think people need to be prepared because next year will be no different. Remember, in 2025, we had a point where we dropped 20%, but ultimately the year could end up rising 20%. So keep in mind that this situation is likely to happen again.

Host: Additionally, what advice do you have for newcomers who entered the market after 2023 and have not experienced a real major correction?

Tom Lee: First, it feels great when the market is rising, but there will be very long periods of pain in the future, and you will question yourself. But it is precisely at that time that you need determination and conviction the most. Because, the money made by investing at the lows is far greater than trying to trade at the highs.

Original link

免责声明:本文章仅代表作者个人观点,不代表本平台的立场和观点。本文章仅供信息分享,不构成对任何人的任何投资建议。用户与作者之间的任何争议,与本平台无关。如网页中刊载的文章或图片涉及侵权,请提供相关的权利证明和身份证明发送邮件到support@aicoin.com,本平台相关工作人员将会进行核查。

Share To
APP

X

Telegram

Facebook

Reddit

CopyLink