BlockBeats News, July 9th. Bank of America Securities pointed out in a research report that Nvidia's current valuation has dropped to the lowest level since the beginning of this AI bull market—its forward P/E ratio is only about 18 times, the lowest in nearly 7 years, compared to the expected P/E ratio of about 21.5 times for the 2027 fiscal year, further dropping to 14.7 times for the 2028 fiscal year. This valuation is not only significantly lower than its historical range but also forms a discount of about 30%-35% compared to the average 22 times/19 times 2027/2028 expected P/E ratio of the five tech giants: Apple, Microsoft, Google, Amazon, and Meta. Bank of America's core assessment is that the current market pricing implies a significant 30%-35% downside expectation for Nvidia's EPS in 2027/2028, a hypothesis that is fundamentally unfounded. From a PEG perspective, Nvidia's 2027 PEG is only 0.3 times, much lower than Apple's 2.7 times, Microsoft's 1.0 times, and Google's 1.9 times. Nvidia's stock price has only risen by 3% so far this year, significantly lagging behind the Philadelphia Semiconductor Index, which has surged by 82%. Bank of America believes that this significant deviation reveals an "enhanced buying opportunity," reiterating a buy rating with a target price of $350, implying over 70% upside from the current price.

Bank of America further dissected the market's two core concerns. Regarding memory cost pressures, Bank of America believes that the market is both overestimating cost pressures and underestimating Nvidia's pricing power— from Blackwell to Vera Rubin, the incremental cost of each rack's HBM is about $200-300,000, but the entire rack's selling price is expected to increase by $2-3 million to $6-7 million, driven not only by memory but also by components like Vera CPU, NVLink, Quantum Ethernet network that do not require HBM, and a range of software functionalities reducing inference costs; Nvidia has built a strong moat through supply chain precommitments of over $119 billion, with gross margin expected to remain in the mid-70% range. As for ASIC competition, Bank of America responded directly with data— since Google TPU's launch in 2015, Nvidia's GPU accelerator sales have increased by about 700 times, with sales to hyperscale cloud providers growing 115% year-over-year, roughly twice the rate of cloud computing capex growth, indicating its continued expanding wallet share among hyperscale customers. Bank of America expects Nvidia to maintain a market share of 65%-70% or more in AI compute capex in the long term, with the upcoming earnings conference call expected to be a key catalyst to clarify its pricing power and moat to the market.