Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
Wow, Buffett put $4.1 billion into a new position in TSM. I haven’t looked at TSM in a long time, absolutely amazing company that I didn’t realize had gotten so cheap. The geopolitical risk involved is large. Anyone here have any thoughts on TSM?? TIA!
Sorry in advance; the following is a bit long.
I spent many years in the semiconductor manufacturing business. TSM is certainly a great company. As an investor, the thing about semiconductor manufacturing is that it is incredibly capital intensive, and it is incredibly difficult. Barriers to entry are enormous. It is entirely dependent on governmental subsidies usually in the form of tax holidays. It requires a tremendous amount of water and electricity.
A factory is called a "Fab," short for semiconductor wafer fabrication plant. The clean room of a fab is the cleanest place on Planet Earth - or outer space, for that matter. It is at least 3 orders of magnitude cleaner than the cleanest hospital operating room (I'm out-of-date, so it could be 4 orders by now). Something as large as a single Coronavirus particle, if it landed on a wafer during manufacturing, would destroy it. Indeed, an errant cosmic ray that gets through and hits a wafer during the months when transistors are "grown" on silicon wafer would damage it. Just one of the tools used inside the clean room is requires TWO Boeing 747 cargo planes to transport. The lead time for those is years.
When you build a fab, you rarely "retrofit" an old one. For a variety of reasons, you start with bare land, and you invest around $10 to $12 Billion to build the fab. That's "Billion" with a "B". It will take you 3 to 4 years before you can begin production, and it takes a while to work out all the kinks so you can produce wafers. It might be 5 years before you generate your first dollar of revenue from that investment.
So, up front, you are making a $10 to $12 Billion "bet" that there will be a market for the chips on the wafers 4 years from now. That's a big bet. And, you're ALSO betting that you can make enough money building those chips so the $10-$12 Billion capital expenditure sees a positive ROI -- before the factory itself is effectively obsolete and the chips it makes are obsolete. You don't shut down an old plant, it continues to operate building chips that are no longer cutting edge, at dramatically reduced prices and almost sold "by the pound". No one cares much about them. It is these old obsolete chips, by the way, that auto manufacturers want for their cars.
Semiconductor manufacturing is heavily dependent on government subsidies usually in the form of tax holidays. Every fab is built with governmental support. And you need a very skilled workforce: the employees in the fabs are highly paid scientists and engineers, typically with PhDs in materials science, electrical engineering, chemical engineering, mechanical, industrial, photolithography, chemistry, solid state physics, etc.
Then, the wafers coming out of the fab are typically shipped to another factory that cuts the wafers into component die, which are packaged into chips, tested & sorted. The Assembly Test plant is much less expensive and doesn't require the PhDs in its operations. As a side note, about once every decade or so, a "perfect wafer" comes off the line. That means every single one of the chips on it actually work. That's a minor miracle. This stuff is only >< this far away from magic. "Yield" is the percentage of chips from the wafer that work; the rest do not work and are pulverized. An average yield of 70% is very good. TSMC they achieved 80% on one of their manufacturing processes, and that is very very good.
Is TSMC worthy of investing in? Who am I to contradict Warren Buffett? The thing is, as an investor you have to understand just what you're signing up for. If you want to invest in semiconductor manufacturing companies, and your eyes are wide open, then yes TSMC is a long term buy.
FND has just laid out another fantastic quarter, with growth plans firmly in place:
+ 25.2% revenue YoY
+ 11.6% same-store sales YoY. 14 years in a row of same-store sales growth!
+ 16.7% adjusted EPS YoY
FND will end the year with 191 warehouses, operating in 36 states. They also have 6 design centers and are planning to increase this service to in-home visits. Their "Pro" business continues to grow on both fronts: more "Pro" accounts, and; higher-ticket sales per order.
I'll continue adding to this position.
Just curious, FIRE42. Most economists say we will enter a recession for real in 2023. By that, I mean a recession as felt by actual consumers in their day-to-day lives. Of course, this year's rate increases also have had a substantial impact on residential real estate transactions, and new owners of a house are one of the sources of customers for FND.
So, buying into FND today seems to be buying into a company that, for macroeconomic reasons, might see fewer and fewer customers during 2023. Of course, this is all hypothetical.
It isn't about FND or its strategy or its operations, its about things beyond their control (economy, layoffs, pain on Main Street, etc).
Do you have any concerns about buying into FND at this stage of the economic cycle?
Separately, I notice "short interest" is over 10% - quite high.
Thanks for the feedback. I am long TXN on the low tech end for exactly the reasons you point out, but with TSM so undervalued, and now that it has Buffett’s blessing, it’s tempting. The geopolitical risk is scary here though, and I usually like to keep my investments here in the good old USA, so I’m learning towards passing on it. Too many considerations make me want to put this one on the “too hard” pile.
I've been an investor in AMD and NVDA for several years. I recommend listening in on the NVDA earnings call later today. Jensen is a genius. Check out their website to read about their cutting-edge technology. And it's not just the hardware, but the software as well that oftentimes is overlooked by analysts. I've mentioned in the past that NVDA is rolling out its first CPU next year, which is geared toward the data-center part of its business. NVDA will be the major player (IMO) in AI for self-driving vehicles as well, so they are more than just a GPU company for gamers. I think you will be impressed after you perform a deep dive into NVDA. And yes, I like AMD just as much!
I've been an investor in AMD and NVDA for several years. I recommend listening in on the NVDA earnings call later today. Jensen is a genius. Check out their website to read about their cutting-edge technology. And it's not just the hardware, but the software as well that oftentimes is overlooked by analysts. I've mentioned in the past that NVDA is rolling out its first CPU next year, which is geared toward the data-center part of its business. NVDA will be the major player (IMO) in AI for self-driving vehicles as well, so they are more than just a GPU company for gamers. I think you will be impressed after you perform a deep dive into NVDA. And yes, I like AMD just as much!
I bought 400 shares if NVDA at $128 a share. I am up roughly 40 points as iof yesterday.
The cost of chips in the newest technologies is no longer dropping. This is a relatively new development. For decades, the miracle of the transistor was that in each new node (the industry term for a transistor of a particular size) the products were better along every axis... speed, power, size, and cost. That's no longer true. If you are interested in reading... do a web search on "the death of moore's law", and you will get various opinions on the subject.
The day will arrive when it's no longer economically viable to make the enormous investment in the next node. As the cost goes up, the potential market shrinks. So investing the many billions in the new technologies has become quite a dangerous game.
Steel huh? Take a look at STZHF, a small cap Canadian steel company, lots of good things happening there. I've been following it for a bit now, still need to do a deep dive on it, but from the outside looking in, it looks like they're making all the right moves.
Just curious, FIRE42. Most economists say we will enter a recession for real in 2023. By that, I mean a recession as felt by actual consumers in their day-to-day lives. Of course, this year's rate increases also have had a substantial impact on residential real estate transactions, and new owners of a house are one of the sources of customers for FND.
So, buying into FND today seems to be buying into a company that, for macroeconomic reasons, might see fewer and fewer customers during 2023. Of course, this is all hypothetical.
It isn't about FND or its strategy or its operations, its about things beyond their control (economy, layoffs, pain on Main Street, etc).
Do you have any concerns about buying into FND at this stage of the economic cycle?
Separately, I notice "short interest" is over 10% - quite high.
Yes, I'm concerned a little bit that my entire portfolio will take a hit if there's a recession. But going to cash isn't an option I'm entertaining, so I'll continue adding to the portfolio. And in full disclosure, I've been an owner of FND since 2019.
My thesis is that Floor & Decor will continue to take commercial and retail business from Home Depot, Lowes, and other smaller players. I think we'll also see the same GFR trend when homeowners transition to remodeling their current home(s) instead of downsizing and/or relocating. HD and LOW took a hit during the GFR, but it turned out to be a great time to add more shares in the long term.
I'm going to continue adding to my position as long as revenue, same-store sales, and new locations continue to grow, albeit, at a slower pace.
If it drops after earnings, I will probably add another 100 shares.
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.
Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
