Aug 26, 2021
[8 min Read]
Currently, according to Utradea's Reddit Sentiment Scanner, $NVDA - Nvidia Corp. is the 5th most mentioned over the past 72 hours, and over the past week (Aug 25th). This information is based off of scanning various NVDA stock discussions to find mentions of NVDA, and analyze the sentiment around NVDA. Majority of the online hype came from news of an NVDA stock split through various NVDA stock discussion boards. The NVDA split ended up being a 4:1 stock split and took place on July 20th, 2021. After the NVDA stock split, their stock continued to rally and the NVDA stock price is currently $218/share.
NVDA is a visual computing company that has operations worldwide. NVDA operates in 2 main segments, which include Graphics, and Computing & Networking.
The products that NVDA manufactures are commonly used in gaming, professional visualization, datacenters, and automotive markets. NVDA sells their products to OEM's, retailers, distributors, service providers, manufacturers, and many other clients.
COVID-19 has ravaged global supply chains causing shortages in many commodities such as lumber, toilet paper, hand sanitizer etc., however, one of the most impactful shortages as of late has been the microchip/semiconductor shortage.
The global semiconductor shortage has caused panic in several industries, but most notably in the automotive manufacturing industry. Many companies like Ford and GM have been forced to halt their production due to this shortage, foregoing hundreds of millions in lost revenue. Although some manufacturers (ie. Tesla) have been able to avoid this shortage, many plants in North America were forced to shut down temporarily, causing higher job loss in the automotive industry. However, the automotive industry is not the only segment of the population that Is struggling for semiconductors and chips.
Due to the recent spike in popularity of cryptocurrencies, many people have turned to mining as a source of income/passive income. However, in order to mine these currencies, miners need to purchase one, if not 100's of GPUs in order to start their mining operation. Just to give you an idea of how many GPUs are demanded by miners, it was estimated that in Q1 of 2021 alone, miners bought 700,000 GPU's, this increased demand has driven the average price for a GPU up 2.5-4x. These miners definitely played a role in the acceleration of the semiconductor shortage and the damaging of global semiconductor supply chains.
Additionally, some analysts believe that PC sales rose by a massive 18.1% due to the pandemic. This is because many people needed personal computers in order to work from home, participate in distance learning, and potentially just for leisure due to the pandemic's constraints on daily life. However, due to the increased demand for GPUs from the cryptocurrency miners, it became increasingly difficult for people to order their own personal PC's. This overall increase in demand from both consumers and miners caused computer and computer part prices to skyrocket, applying further pressure on global supply chains.
Due to the huge demand from these separate industries/populations, semiconductor and chip manufacturers like NVDA, currently have a large opportunity for once-in-a-lifetime sales/profits.In a recent market study, analysts have estimated that the global semiconductor industry is set to grow at a CAGR of 8.6% over the next 7 years (until 2028). This helps investors to recognize the lasting demand for these semiconductors, and that it may be worth holding for the long run. Furthermore, analysts are also forecasting the global GPU market to grow at a CAGR of 33.6% for the next 6 years (until 2027). This forecasted growth is great news for semiconductor/GPU companies such as NVDA.
As previously mentioned, NVDA has 2 main segments to their business, which are Graphics, and Computing & Networking. These segments will be broken down and explained in this section.
My comparable analysis requires 4 companies, in which I can compare NVDA's financial ratios, to the ratios of their 4 biggest competitors.
The 4 closest competitors that I found were Intel, AMD, Micron Technology, and Taiwan Semiconductors.
I chose these 4 companies given their market caps, their operations, their geographies, and their business models.
$INTC - Intel Corp: Intel designs, manufactures, and sells their technologies to their various customers worldwide. Intel offers a wide variety of computer parts such as CPU's, GPU's, chipsets, memory, and storage solutions. Intel also has machine learning and AI projects on the go.
$AMD - Advanced Micro Devices Inc: AMD is a global semiconductor business that operates in 2 segments, which are Computing and Graphics. AMD's graphics segment consists of their GPU revenues, and their Computing segment consists of accelerated processing units, chipsets, and development services.
$TSM - Taiwan Semiconductor Manufacturing Co: TSM manufactures and sells their integrated circuits and semiconductors, to their customers across the globe. TSM also offers other services like customer service, account management, and engineering services.
$MU - Micron Technology Inc: Micron Technologies designs, manufactures, and sells memory and storage products to their customers all over the world.
I used my own WACC model to predict NVDA's WACC to be 7.35%.
I was able to find my CAGR of 30.80% by taking the average analyst growth forecast for NVDA over the next 5 years. I used this figure as a constant growth rate over the first 5 years of the DCF model.
Risk Free Rate (Perpetual Growth):
I was able to find NVDA's risk free rate on a website called Finbox. Finbox estimated NVDA's risk free rate to be 2.5%. I used this as my perpetual growth rate for 2030 and beyond. Furthermore, from 2025 to 2030 I slowly tapered down the CAGR to meet this growth rate in 2030.
Operating Expense Increase Rate:
By using the historical growth figures, I forecasted NVDA's operating expense figure to increase by 26.3% between 2021-2025, and then I tapered this growth rate down to 11% come 2030.
Interest Expense Increase Rate:
I was able to find this figure by taking NVDA's yearly interest expense decrease over the past 5 years and average it to find an interest expense increase rate of 37%.
Depreciation and Amortization Increase Rate:
I was able to forecast this figure by taking NVDA's historical depreciation and amortization figures over the past 5 years. By doing this I arrived at an increase rate of 5.1%.
CAPEX Growth Rate:
I was able to find NVDA's CAPEX growth rate to be 9.7%, by using their historical CAPEX to forecast their future CAPEX.
I found NVDA's effective tax rate to be 1.7% through their SEC 10-K filing.
In order to value NVDA, I decided to undergo 3 comparable analyses, as well as 2 different DCF models. I did this in the hopes of achieving unbiased, well-rounded results, to show multiple cases (bullish, and bearish)
I was able to conduct my DCF model by using the information found above in the “valuation information” section of this report. My DCF model estimates that the fair value of NVDA should be $269/share, which would result in a share price increase of 24%. This is a reasonable estimate, however, I decided to undergo 3 comparable analyses in hopes of achieving consensus.
By comparing NVDA's P/B ratio to that of their competitors (listed above in the “competitors” section), I found NVDA's fair value to be $140/share. If this were the case then the downside of this investment would be 36%, this would be a large downside given the potential that NVDA has, so I decided to undergo further comparable to see if this valuation was consistent.
By comparing NVDA's EV/Revenue multiple to that of their competitors, I arrived at a fair value of $70/share. If this was the case, then there would be an implied downside of 68%. This result reaches the same conclusion as reached in the P/B comparable (NVDA is overvalued), however this result is more drastic.
By comparing NVDA's PEG Ratio to that of their competitors, I arrived at a fair value of $185/share, which implies a downside of 15%. This implies that NVDA is slightly overvalued, however, it is not as overvalued as some of the other comparable analyses estimate it to be. As a result of the variance in price targets via the 3 comparable analyses, I decided to undergo a weighted average for my results.
For my weighted average comparable, I decided to split the weight 40/40/20, for P/B, PEG, and EV/Revenue respectively. By doing this I arrived at a fair value of $144/share, which implies a downside risk of 33%.
In order to put one final, all-encompassing, price target on NVDA, I took a weighted average between the DCF valuation and the average comparable valuation. This average was split 75/25 for the DCF and average comparable valuation(s) respectively.
By doing this, I arrived at a price target of $237.50/share, which implies a 9.3% upside.