Loftus Peak is one of the big local winners in the AI boom.
The ASX-listed Global Disruption Fund posted a staggering 35.3% return for the year ending June 30, and a staggering 42.5% for the first six months of the calendar year.
The company’s top holdings include many of the brightest mega-cap stocks, including Nvidia (up 195% year-to-date), Amazon (up 48%), Alphabet (up 31%) and Microsoft (up 38%). It is included. ) and chipmaker AMD (up 77%).
hot questions remain
It was an astounding growth, but Polak is well aware of the compelling questions facing investors after this phenomenal rise. The question is, has this AI stock boom gone too far, or too fast?
Pollack isn’t surprised by the size of profits since the beginning of the year. Loftus Peak’s long-term bet is in a world that needs to manipulate ever-increasing amounts of data at ever-higher speeds, and few technologies align these trends better than generative AI.
But the speed of the rally is shocking, and investors shouldn’t necessarily expect a repeat of the rally seen in 2023, he said.
“Our model claims this should have happened. But our model doesn’t say it shouldn’t have happened so quickly,” Pollack says. “I think it will continue to grow from here.”
The question of whether AI is overhyped or real is so important today for two reasons.
The first is that investors in Japan and the United States will soon enter the earnings season, and AI will be one of the hottest topics. Tech companies clearly have a lot of hype to live up to, but corporate investors across the economy will be looking for evidence that companies aren’t missing out on this potential wave of innovation.
The second problem is that the surge in AI stocks has made valuations look expensive, especially in the United States. Not only are AI leaders up 25% relative to the overall U.S. market, but the concentration in the index is close to all-time highs, and the tradition between stock multiples and real bond yields (i.e., inflation-adjusted) relationship is broken.
Lisa Charlett, chief investment officer at Morgan Stanley Wealth Management, said the last time real yields on 10-year U.S. Treasuries reached their current level of 1.8% was in October 2022, when the U.S. He pointed out that it was the highest yield on government bonds.&The P500 index was at bear market lows, trading at 17.3 times expected earnings. The market is now up 23% and trading at 20 times forward earnings, which “seems totally unjustified” given that bond yields are rising instead of falling, Shalett said. claims.
Goldman Sachs has a different view. AI winners don’t trade at extreme earnings multiples, said Eric Sheridan, who heads internet and software companies at Wall Street firm David Kostin, a U.S. equity strategist. says it sees an equity risk premium and long-term earnings growth at the index level. Forecasts are roughly in line with historical averages.
However, Goldman’s experts believe that previous technology-driven productivity booms, such as the spread of electricity in 1919-1929 and the Internet boom in 1995-2005, created a stock market bubble that eventually burst. I am wary of the fact that
valuable pockets remain
Kostin says it’s important to remember that even if an AI company can generate significant revenue growth over the next few years, that doesn’t necessarily guarantee a higher share price. Most tech companies had significant sales growth from 2000 to 2002, but still his PE multiple fell by more than 50% and his stock price plummeted.
While market watchers continue to argue that AI stocks are ripe for a pullback, Pollack said the current bull market in AI stocks was not without alarming moments. AMD shares plunged in May on concerns about the near-term outlook, while Nvidia shares fell more than 6% in late May and more than 7% in late June. The stock is up more than 5% over the past six weeks.
Pollack expects the current surge in AI-related activity to continue over the next 12 to 18 months, with most major companies looking to explore how generative AI can be introduced into their businesses. Claims it will cost.
“It’s a competitive advantage, but competitive advantage isn’t something you get every 12 months.
Pollack can imagine spending stopping within 18 months, but in the meantime there is still AI value to be harnessed, he said.
Chip makers and other gadget makers such as switches should do well (Loftus Peak’s holdings as of June 30 include Nvidia, AMD, and Marvell Technology) and bring AI to big companies. Companies that can provide may have a continuing (albeit more modest) profit. valid application.
Pollack said Morgan Stanley’s recent analysis looms over Microsoft as the big winner, given that it can effectively change the premium for AI-enabled versions of applications it already sells between 100% and 500%. agree to Assuming he is 20% hired across Microsoft’s existing customer base, that could add another $19 billion ($28 billion) in fiscal 2025. Morgan Stanley has set a target of $415 per share for Microsoft’s current stock price of $332.
Morgan Stanley’s analysis is an example of how the AI narrative is “not built on hype, but on a set of applications that support that hype,” Pollack said. said. So companies like Microsoft aren’t monetizing eyeballs or Internet traffic, they’re monetizing actual products that customers already appreciate.
But Microsoft’s example raises another question. If it becomes easy for everyone to adopt AI, the competitive advantage it provides will be undermined. “The investor risk that worries me most is that generative AI technology becomes so ubiquitous that it becomes a commodity,” said Goldman analyst Kash Langan. If not, how can a company charge a premium for it or monetize it?”
This is a great little example of the questions investors need to ask themselves about AI. The hype is woven in, but what is the reality worth?
Even longtime AI believers like Alex Pollack say it’s useful to remember how the investor flock works. “That’s how markets go … nothing, nothing, nothing, then bang. And they work the same way vice versa.”
