You are currently viewing SemiWiki as a guest which gives you limited access to the site. To view blog comments and experience other SemiWiki features you must be a registered member. Registration is fast, simple, and absolutely free so please, join our community today!
The ranking of largest semiconductor companies by market capitalization are changing drastically in the past 12 months. Will Qualcomm, AMD, or Texas Instruments move upward to pass Intel in the near future?
The market capitalization as of Friday, November 12, 2021 is follows:
If the AMD merger is approved, it will be larger than Intel in market cap by the end of this year, if not I still think it can happen in 2022. Either QCOM or TXN in 2022 is also possible.
I don't think IDM 2.0 will be successful, and my belief is that until Intel decides to commit to the Fabless model it's slide will continue. Intel competes too broadly with it's potential IFS customer base and the frienemy model will not work, especially not if you are a former monopolist who has a reputation for using dirty tricks at every turn.
PREVIOUS CLOSE
The last closing price
$303.90
YEAR RANGE
The difference between the high and low prices over the past 52 weeks
$115.67 - $323.10
MARKET CAP
A valuation method that multiplies the price of a company's stock by the total number of outstanding shares.
759.75B USD
VOLUME
The average number of shares traded each day over the past 30 days
33.55M
P/E RATIO
The ratio of current share price to trailing twelve month EPS that signals if the price is high or low compared to other stocks
108.59
DIVIDEND YIELD
The ratio of annual dividend to current share price that estimates the dividend return of a stock
0.05%
PRIMARY EXCHANGE
Listed exchange for this security
NASDAQ
PREVIOUS CLOSE
The last closing price
$303.90
YEAR RANGE
The difference between the high and low prices over the past 52 weeks
$115.67 - $323.10
MARKET CAP
A valuation method that multiplies the price of a company's stock by the total number of outstanding shares.
759.75B USD
VOLUME
The average number of shares traded each day over the past 30 days
33.55M
P/E RATIO
The ratio of current share price to trailing twelve month EPS that signals if the price is high or low compared to other stocks
108.59
DIVIDEND YIELD
The ratio of annual dividend to current share price that estimates the dividend return of a stock
0.05%
PRIMARY EXCHANGE
Listed exchange for this security
NASDAQ
I think that Nvidia is hugely overpriced. For last 5-7 years the GPU market hasn't been that competetive and Nvidia was able to exploit that and massively increase prices and production. But the competition is already heating up and Nvidia doesn't have cpu's like AMD and Intel so the company will be at disadvantage.
I think that Nvidia is hugely overpriced. For last 5-7 years the GPU market hasn't been that competetive and Nvidia was able to exploit that and massively increase prices and production. But the competition is already heating up and Nvidia doesn't have cpu's like AMD and Intel so the company will be at disadvantage.
I made the error of omission, dismissing NVidia as overpriced for years and missing out on the big run. NVidia at $700 million market cap and 100x P/E ratio does not make sense to me either, but it's still growing fast and should not be dismissed so easily. That said, AMD looks like a bargain to me even at the current price.
I made the error of omission, dismissing NVidia as overpriced for years and missing out on the big run. NVidia at $700 million market cap and 100x P/E ratio does not make sense to me either, but it's still growing fast and should not be dismissed so easily. That said, AMD looks like a bargain to me even at the current price.
If you look at Intel it also looks like bargain. But short-to midterm they will be stagnating, Nvidia will be growing but I don't think they have the technical leadership they had for years. So the growth will potentially slowdown. That same can happen to AMD but they are quickly acquiring new customers in many segments.
If you look at Intel it also looks like bargain. But short-to midterm they will be stagnating, Nvidia will be growing but I don't think they have the technical leadership they had for years. So the growth will potentially slowdown. That same can happen to AMD but they are quickly acquiring new customers in many segments.
Intel could be a bargain if it changes strategy. IDM 2.0 will not work. If Intel can spin off IFS and become a fabless company, I believe it will be the start of a big turnaround in the company and stock price.
Most of these companies have meme stock type P/E ratios. Intel does not. For example, NVIDIA has an order of magnitude higher P/E than Intel. It may last for a while but at some point the music will stop.
Most of these companies have meme stock type P/E ratios. Intel does not. For example, NVIDIA has an order of magnitude higher P/E than Intel. It may last for a while but at some point the music will stop.
The executives at those super high P/E companies know it is not sustainable. That's part of the reason they are buying suitable companies to improve it. AMD is buying Xilinx and Nvidia is buying ARM. It will not solve the problem totally but at least it will help.
The executives at those super high P/E companies know it is not sustainable. That's part of the reason they are buying suitable companies to improve it. AMD is buying Xilinx and Nvidia is buying ARM. It will not solve the problem totally but at least it will help.
AMD P/E of 45 is actually very cheap for a company growing 50%+ a year. Nothing high about QCOM's 21x P/E ratio. Broadcom's P/E ratio is a bit high, but Hock Tan has proven himself over and over again and investors have confidence that he will be able to continue the growth. NVidia P/E is very high even considering it's growth, but the stock could fall 50% tomorrow and still be 50% larger than Intel by market cap.
The fact that these companies are being valued more than Intel is not an accident of the market in my opinion. They reflect the very strong fundamentals of these market leaders vs the very real challenges Intel is facing. And if Intel reclaims it's leadership spot, it won't be because AMD/NVidia/Qualcomm/Broadcom faltered - it'll happen if either IDM 2.0 is successful (not likely in my opinion) or the company goes fabless.
When you look at these market caps and divide them by the annual revenues and then consider the final product margin (extremely high for Intel historically), there seem only two factors left to explain why Intel is an outlier:
1) growth of the markets served - the market may be over-estimating this for the others and there may well still be growth in x86/server/data center
2) the fact that Intel has fabs and the others are fabless
While I'm rather sceptical of these extremely high market caps ... depending on how you weight these two factors, the market might arguably be saying that Intel would do better if it were fabless and that the fab side of the business may be of no - or even negative - value. If Intel's fab cost is higher than TSMC's (no public data though there seems to be a general consensus on this) and if TSMC could provide a suitable process (I don't know - but if AMD use them, it seems possible), couldn't Intel's margin actually improve by using TSMC ?
Just a thought that came into my mind while reading above. Interested to hear other views and insights.
When you look at these market caps and divide them by the annual revenues and then consider the final product margin (extremely high for Intel historically), there seem only two factors left to explain why Intel is an outlier:
1) growth of the markets served - the market may be over-estimating this for the others and there may well still be growth in x86/server/data center
2) the fact that Intel has fabs and the others are fabless
While I'm rather sceptical of these extremely high market caps ... depending on how you weight these two factors, the market might arguably be saying that Intel would do better if it were fabless and that the fab side of the business may be of no - or even negative - value. If Intel's fab cost is higher than TSMC's (no public data though there seems to be a general consensus on this) and if TSMC could provide a suitable process (I don't know - but if AMD use them, it seems possible), couldn't Intel's margin actually improve by using TSMC ?
Just a thought that came into my mind while reading above. Interested to hear other views and insights.
Well yes, Intel's position would improve dramatically if they went fabless(which off course won't most likely happen). But their biggest problem is bad management and lack of talent. Pat called some of his old friends into the company, with all due respect to these highly accomplished people I don't think they will be able to compete with younger people that are in AMD and Nvidia.
While I'm rather sceptical of these extremely high market caps ... depending on how you weight these two factors, the market might arguably be saying that Intel would do better if it were fabless and that the fab side of the business may be of no - or even negative - value. If Intel's fab cost is higher than TSMC's (no public data though there seems to be a general consensus on this) and if TSMC could provide a suitable process (I don't know - but if AMD use them, it seems possible), couldn't Intel's margin actually improve by using TSMC ?
Just a thought that came into my mind while reading above. Interested to hear other views and insights.
I completely agree with this assessment. Intels foundry business, in my opinion, is a tremendous drag on the company. They will need to spend $100b over the next 3 years to match TSMC capex, and even then they will be a higher cost operation and still won’t be anywhere close to matching TSMCs ecosystem.
On the point of AMD, AMD margins are trending up, and will probably be higher than Intel margins in 2022. So yes, I believe a fabless Intel would also have much higher margins.
Food for thought. Intel is trading at 10x earnings, but every dollar of those earnings and then some need to be spent on capex to stay relevant as a foundry. In fact, you can expect negative $10b in free cash flow every year for the next 3-5 years. Even then, the odds of Intel actually succeeding with IFS is less than 50%, and if they fail they will have burned tens of billions on an unprofitable business that will continue to demand billions a year in capex. Compare that to Qualcomm, where 80% of income goes to free cash flow. Intel is cheaper on earnings but which company is cheaper on a discounted cash flow basis?
It should be obvious that Intel would be better off as a fabless company but there are big egos at play.
GPU maybe more important than the cpu. That sort of environment favors tsmc. Nenni's idea of splitting manufacturing with tsmc is probably the right way to go because Intel could use tsmc to collaborate with nvidia, Marvell and other ip and make the soc or chiplet themselves.
Intel is firing managers they perceive as dead weight.
At current price, Intel is valued at $204.84 billion and AMD is at $186.24 billion according to Google. There is a chance that AMD's market capitalization may surpass Intel's before AMD completing Xilinx acquisition.
Well yes, Intel's position would improve dramatically if they went fabless(which off course won't most likely happen). But their biggest problem is bad management and lack of talent. Pat called some of his old friends into the company, with all due respect to these highly accomplished people I don't think they will be able to compete with younger people that are in AMD and Nvidia.
During a recent speech in Taiwan (see link below) Morris Chang recalled when he just started TSMC around 1987, an old colleague from Texas Instruments visited him. That person advised Morris that it can lead him to failure if Morris doesn't bring his old team to TSMC. That really irritated Morris.
Morris explained that he believes bringing people who worked with you before to an organization is not a wise practice. He thinks it can create fractions within the organization and limit the ideas, visibility, and options a leader should have. He said at most he only brought his previous secretary to a new business division or a new company.
Granted every company has its own environment and challenges, but Morris Chang's management philosophy and practice is unique.
The speech was about how Morris Chang grow and learn. But there was a Q&A in the end. I will summarize Q &A and update late．． 1. He knows every Intel CEO including Pat. Pat visited Morris in Taiwan in 2015. Morris was impressed that Pat can explain what VMware is very clear in 15 mins. In...
