Comments on Nvidia’s Q4 : Standing on the Edge of AI, Be Wary of High Valuations.

AI chip manufacturer NVIDIA (NVIDIA) announced the fourth fiscal quarter of fiscal year 2023 and the financial report for fiscal year 2023 as of January 29, 2023, as well as the performance guidance for the new fiscal year after the U.S. stock market closed on February 22.

Revenue in the fourth fiscal quarter of fiscal year 2023 was US$6.05 billion, down 21% year-on-year and up 2% quarter-on-quarter, and net profit was US$1.4 billion, down 53% year-on-year, exceeding market expectations. Non-GAAP diluted earnings per share fell 33% year-on-year (52% quarter-on-quarter) to US$0.88, and non-GAAP gross profit margin fell 0.9 percentage points year-on-year (10 percentage points quarter-on-quarter) to 66.1%.

Inventory turnover days increased to 211.66, YoY+136.08%, which is expected to be close to the high point of inventory. Although the inventory is high, the company's gross profit margin has not declined significantly, mainly because the company uses the strategy of increasing the price of new products to force consumers to buy inventory products, and does not use methods such as price reductions to clear inventory, but focuses more on new products. Nvidia released a new RTX 40 series graphics card in the middle and late last year. The performance improvement is limited but the price has been significantly increased, forcing consumers to buy old graphics cards. It can be seen from the management's statement that "the inventory crisis of game graphics cards has basically ended" can also be seen that the inventory management of graphics cards has been quite effective.

According to the revenue terminal market segment, Nvidia's main revenue comes from the data center (Datacenter) and gaming (Gaming) business, which accounts for about 90% of the company's revenue. The remaining sources of income are Professional Visualization, OEM&IP, and autonomous driving.

The data center business revenue was US$3.616 billion, YoY+10.82%, and the revenue growth rate dropped sharply, from 83.11% in 22Q1 to 10.82%, mainly due to the reduction of capital expenditure by large cloud computing manufacturers and the purchase of data center chips. It is expected that the data center business will still be under pressure in the short term. In the long run, with the increasing demand for cloud computing, the data center business will still be able to return to high growth.

The revenue of the game business was US$1.831 billion, a year-on-year decrease of 46.46%, dragging down the company's overall performance. The slump in the game market is mainly due to the fact that after the epidemic, the duration of consumer games has become shorter, and dealers stocked up more graphics cards in the early stage, so there is insufficient demand. At the same time, the slump in digital currency (especially Ethereum) has also led to a lack of motivation for mine owners to purchase graphics cards for mining.

In terms of other businesses, the visualization business revenue was 226 million US dollars, a year-on-year decrease of 64.85%, accounting for 3.73% of the revenue; the autonomous driving business revenue was 294 million US dollars, a year-on-year increase of 135.2%, and the revenue share increased to 4.86%; OEM&IP business Revenue was US$73 million, a year-on-year decrease of 68.8%, mainly benefiting from the growth of the autonomous driving market, including autonomous driving and artificial intelligence cockpit solutions.

As the game business is expected to bottom out and AI’s demand for computing power increases, Nvidia has given an estimate of $6.5 billion in revenue for the new quarter, higher than Wall Street’s previous estimate of $6.2 billion.

This earnings call will be more like Nvidia's AI presentation. According to a user on Twitter, Nvidia mentioned AI 23 times in five minutes. The AI wave set off by ChatGPT has concealed Nvidia's relatively weak financial report to a certain extent, but the demand for AI may not be reflected in the company's performance until the second half of this year.

Nvidia's leading position in the AI field comes not only from powerful hardware chips, but also from the software ecosystem. On the basis of GPU hardware, Nvidia has derived a CUDA software ecosystem based on general computing. The rapid growth of the CUDA ecosystem has enabled NVIDIA to have a dual scale effect of software and hardware. CUDA is a general-purpose parallel computing architecture launched by Nvidia. This architecture enables GPUs to solve complex computing problems, and gradually has the ability to compete with CPUs.

In order to help enterprises enter the AI field, Nvidia has also launched its AI cloud service products. At present, Nvidia has reached a cooperation with Oracle Cloud. Users can access Nvidia DGX AI supercomputers through Nvidia DGX Cloud on Oracle Cloud Infrastructure. DGX Cloud will drive demand for cloud service providers, as Nvidia has many customers, including AI startups and enterprises, who want to run AI applications in the cloud, in hybrid cloud configurations, or in multiple cloud services.

The AI story is a good one, and it makes sense in the long run. However, Nvidia's recent surge has priced in this part of the expectation. From the perspective of actual performance, it is difficult for the data center business to return to high growth in the short term, and the recovery of the game business is limited, so the performance growth of the entire company is relatively slow.

The fundamentals of such a slowdown in performance growth are difficult to support the current high valuation in the face of rising terminal interest rates.

How high is Nvidia's valuation? You can take a look at the following commonly used indicators as a reference. Whether it is from PB, EV/Revenue (Forward), EV/EBITDA (Forward), or EV/CFO, Nvidia’s valuation is much higher than that of technology peers and competitors ( Tesla, Qualcomm, Ford).

Therefore, in terms of trading strategy, it is expected that Nvidia's safe range is below $150.It is not recommended for everyone to chase up and buy stocks;Options can be considered short, buy a put with a higher strike price, or sell a call with a lower strike price.