Memory Is The New Bottleneck

Straight Talk for All - Nonsense for None

Please check out our other podcasts:

https://skepticsguidetoinvesting.buzzsprout.com

Disclaimer - These podcasts are not intended as investment advice. Individuals please consult your own investment, tax and legal advisors. They provide these insights for educational purposes only.

Transcript

Welcome And The Memory Thesis

Clem Miller 0:02

Hello, everybody, , and welcome to Skeptic's Guide to Investing. This is Clem Miller, fresh back from a trip to the Great North of Canada, especially Newfoundland and the land of icebergs and puffins and whales and and and fried food. Uh and and today we're gonna discuss a number of topics, but on this on this particular episode, we're going to talk about memory, not not our failing memories, personally, , but the issue of memory space, you know, memory chips, AI, data centers, and , and so on. So um, and of course, I'm Clem Miller, and I'm here with Steve Davenport to talk about this important issue and how it might affect what's going on with markets. So, Steve, my my first question to you is you know, we've seen seemingly an increasing amount of volatility from semiconductor and data center support type stocks. Uh one day they're up a lot, the next day they're down a lot, the next day they're up a lot. Uh you know, it's hard to, you know, certainly in investing you should never make decisions on a day-to-day basis. And this volatility is validation of that, right? But the question is, are we seeing a trend now where the the sort of seemingly inexorable upside surge in these stocks has been halted or thwarted by limitations such as power and so on. So so Steve, whatever, you know, what what do you think about all this?

Steve Davenport 2:00

Well, Clem, I think it's one of the more interesting developments because I'm a value guy as you're more of a growth guy. And I looked at you know micron when it had its great move, you know, two weeks ago, and everybody started to say micron is the new black. And everybody was you know looking at micron and seeing their earnings estimates go from $25 to $250. So when you see earnings go up at a 10 times ratio, you've got to ask yourself, is this sustainable? Is this real? Is this something that we can understand? And then you look at the the actual um PE and memory has been a commodity. Memory has been the last thing you think about when you think about building servers or building of data centers. And now because everybody's interested in the speed of the memory and the the ability of the memory to um help solve these complex problems, you realize that having a CPU or GPU that you know does the processing and tries to solve deeper and deeper questions, well, they're gonna need to access more and more memory

Why AI Needs Faster Memory

Steve Davenport 3:20

because they're gonna look at the problem and break it down into sub-problems. And those sub-problems are gonna get stored somewhere as they're being solved, and you know, the memory piece of this and the movement between the chips is gonna become more and more critical. And so what I think is interesting today, as we look at you know, the earnings for Micron or after the bell, I think we're gonna get a lot of information as to one is you know, is there a new advantage to their technology in memory? Um, or do other players like Samsung and SK Heinex, you know, can they catch up? Can they be, you know, the microns instead of um, you know, in second place? And is the technology in memory something that's gonna, you know, differentiate people? And just like we have, you know, NVIDIA is the best for some of their calculation speeds and some of the you know their chips set, and we'll have a you know, we'll have the gold standard, the silver standard, and the bronze standard. And I think it's one of those things that when we look at this, I don't think we ever expected the AI conversations would also be conversations about energy. And whether we use nuclear or whether we use our local energy source, I mean, I think what is AI is doing is breaking it down into several different problems from an investment standpoint. One, do you have the chips? Well, if the chips don't have energy to drive these um analysis and AI technology, then this that it stops. And we have to figure out you know, do we get our own energy source, which is you know what I would call the the remote nuclear or a small nuclear, um, which is a whole industry in itself. And I look at memory and I think this has become its own, you know, specialty. And so we used to just say chips, and we were generally describing anything that goes on a board. And now I think we're we're talking about power and how power storage goes at the chip level. And um, I guess I would say it's it's becoming a lot less vanilla. Um, and chips were chips, and um, we all were happy that we understood that chips were important to technology. And now I'd say we have to get a you know a degree in chip technology before we can figure out what to invest in. I don't know if I can differentiate between the memory capability and speeds of you know Heinek, Samsung, and and Micron. So how do we how do we invest in a space that seems to just get more and more complicated? And the one thing I say, Clem, is

Power Limits And Data Center Reality

Steve Davenport 6:30

keep it simple, stupid. Um, and and then for me, when I look at Micron and I see even after you know this meteoric rise, it's still trading around 11 or 12 times earnings. And so being , you know, C B or I don't know if you've known the term, but it's a cheap bastard, I I look at things and think, you know, why wouldn't I invest in this as my exposure to AI at a multiple of 11 or 12 versus an average of 40 or 50 for some of the other you know AI number chips and exposures? I think there's less downside when you're at a 9 to 11 than you are in a 40 to 50, 60. And now with SpaceX, we're not even talking about earnings, we're talking about 100 times revenue. So therefore, I I look at that as an extreme climb. And I say, is the opportunity in micron at 12 times, you know, let's be conservative, or is the opportunity at SpaceX at, you know, now it's gone from 110, 120, it's back to below the price of the IPO, somewhere, you know, in that um 100 times um revenue. And how do you even, you know, do we even put SpaceX in the discussion until they get earnings? I mean, I how important are earnings to you, really? I mean, do you really care about earnings or do you really just want to get involved in technology and understand that you have the greatest, you know, one of the greatest names with the one of the greatest leaders and and inventors in in history with Musk? I mean, do you just forego earnings?

Clem Miller 8:22

Well, I think I think we're gonna talk about Musk and SpaceX in another podcast.

Steve Davenport 8:29

Right, but you could see where it overlaps a little as an investor, right? Okay, well, I mean, the the question is should does it being 10 times earnings make you feel better as an investor? Uh relative to SpaceX, so to speak. Relative to any relative to 40 times for net video.

Clem Miller 8:49

So I I would point out that yeah, I mean, 10 times earnings is very good, right? Uh, but I would it's even better if you look at peg, right? So I was gonna I was gonna get there. Yeah, the forward peg for micron, since you mentioned micron, is actually 0.10. Which is which is insanely low. Correct. Which which makes me believe that there's a lot of upside, but also makes me question the earnings forecasts.

Steve Davenport 9:26

Well, that's what I don't know how that's what I want to get into with you is your pairs always depend on that forward view, yeah. You always believe that analysts are gonna be you know touting their particular favorites with the those numbers. I I think those forward earning numbers, I mean, I always look at consolidated forward estimates, meaning that I like to not just see a spread of them. I like to see, okay, who are the best analysts and what's the average of those best analysts versus you know, because I think that forward earnings is where we get in trouble when we start going two years, three years,

Valuation Debate Around Micron

Steve Davenport 10:05

four years, or five years forward when we don't know what the next quarter is going to bring, right?

Clem Miller 10:11

Now, offsetting Peg, okay, what I've noticed, because you know, I follow short interest, right? Right noticed is that short interest has gone up a lot for a lot of these stocks. They used to be used to be a lot lower, okay, and now it's gone up a lot. So there are people out there, short sellers out there, who are betting against these stocks. Yeah, I think that that's healthy, isn't it? Oh, it's healthy, and but it also tells me that I wouldn't get um I wouldn't get overconfident about these stocks. Let me put it that way.

Steve Davenport 10:51

My question is an overconfidence club. My question is we have investors out there who want to be part of AI. They didn't get into everything as much as they wanted to, and now you see this focus on memory. And my question to you is, is this the big opportunity for those who have been left behind to get into memory because memory is a lower multiple and also going to participate very widely in this AI future? So if you were somebody who's sitting at home saying, Boy, I really missed it and blew it with um, you know, not getting into more cloud to cloud and and and other technologies, is is this their last chance to get on the bus? Or is there gonna be another chance or you know, nuclear-powered units that are gonna fund, you know?

Clem Miller 11:48

Well, let's let's let's be yeah. Nuclear power is something that takes a long time to develop, get regulated, get you know, regulatory licensing, and so on and so forth. So even for the so-called micro power or small powered units, it's gonna take a long time.

Steve Davenport 12:08

So I won't divert the let's not divert it, let's just say is it is memory an opportunity for you if you were somebody who's weren't able to participate in the AI rise thus far. Of course.

Clem Miller 12:24

Why didn't we look at it? I think there's still I think there's still an opportunity, and , and I'm involved. Uh I've got a lot, you know, it depends on what you're talking about as a memory, as memory, right? But there's semiconductors in there and other things, right? Uh, but I I would I mean I've got a number of stocks in my portfolio that broadly speaking are um you know in this group. So you know, I've got Taiwan semiconductor, right? Uh I've got where else? What else do I have? I mean, what percentage of their business micron, lamb research, right, fit comfort systems, Celestica, ASML, applied materials.

Steve Davenport 13:10

Yeah, but those are all they're the majority of their business is in memory except for micron. Yeah, yeah. So what I'm saying is, do you want a focused exposure to memory, or is memory a commodity? That's the way it's been treated for years, is it it always has a lower multiple, it always is cheaper, and everybody says, yeah, it's cheap for a reason because you could basically get a memory anywhere, and it's not, you know, there's there's not a lot of technology in memory. So therefore, it's never gonna get the the multiple and the, you know. I mean, that's this historically that's the way memory has been thought of, right? Yeah, it pays a dividend, it's it's very regular, it's very dependable, but now that you're gonna need exponentially more memory, does the the ones who do it best, like micron, do they get a premium and do we start to look at them differently? Because now memory is really critical to solving these problems.

Clem Miller 14:16

I agree. I mean, I I've got Micron and I'm not selling it. Okay. Um you know, I I used to have, I don't know if if you would consider Sandisk to be part of that team. I think Sandisk is one of the names that's I had I had Sandisc, and and it's still cheap by the forward peg measure, but I saw it go up a lot over a a few a couple day period, which gets nervous. And so I I ended up selling it maybe too early, but you know, I I'm being over the last six months, eight months, ten months,

Short Interest And Selling Discipline

Clem Miller 14:55

I've been very careful about about cashing in on stocks that have risen a lot. So um, which is why I have a lot of cash in my portfolio.

Steve Davenport 15:05

Why did why do you sell Sandus versus Micron? Because they've both gone up a lot.

Clem Miller 15:11

They've both gone up, but say so typically to share an insight into my sales to sale discipline, and there's nothing terribly dramatically insightful about this, but when I see something rise by more than 10% in one day, that leads me to question whether we're going to see more of that or not. And it makes me doubtful as to whether we're gonna see more of that or not. So I tend to sell a part or all of something when I see a when I see something like that surge. So I I remember, I don't know if you remember this. Uh there was a time last year when there was a one-day surge in Oracle. Yeah, 33%. Yeah. And so so was I gonna hold on to that? No freaking way. I sold that, right? Uh, and pocketed the profit. So that's what I tend to do. So I tend to look, I mean, , if it's under 10%, I don't worry about it, right? But if it's over 10%, then I I look into it and then think about whether I want to sell that. And I I by 10, I mean 10 in a day, right? But now, so you know, my micron has been more stable and steady in its increases, it's increased a lot, but I don't think I've seen it increase by more than 10 in a day.

Steve Davenport 16:38

Um, it had some day a little a little while ago that it was it went crazy, but I may have been away for that. Yeah, I think it's my my point is is that it's a it's a it's an interesting space because I I want to get back to the multiple again. The pegs matter in my mind. And when you're when you're looking to get exposure to something, I mean we can't put all of our money in memory at nine to eleven times and hope we're gonna get the returns of AI, right? Yeah, but we should really think about it as we think that it's stretched and we're worried. And I guess what I'm thinking, and this is just my tactic, is the market is looking at things a little tougher, and the the scare of of the selling in Korea really caused, you know, some some profit taking yesterday. And then if Micron doesn't say all the right things today, and we know that they're gonna have great numbers, but I think the number that everybody's gonna ask about is what's the forward? What's the forward number and what's the forward estimates, and how good do you think you're gonna be and able to achieve it? If they stumble today and we see a pullback in the 10 or 20 percent range, is that gonna be a big opportunity for our investors? Or is it gonna be, you know, how do you how do you look at that stumble?

Clem Miller 18:10

So, Steve, so this is something I have to grapple with today, right? I mean, I've we've seen a little bit of a pullback in Micron the last few days. Uh I think honestly, I think it's anybody's guess as to what's gonna happen tonight, right? With micron.

Steve Davenport 18:27

Correct. All I'm saying is opportunity seems to be knocking. Are you gonna answer?

Clem Miller 18:32

Well, do I want to be do I want to be conservative and pocket the pocket the profits I've made off Micron over the last few months? Uh or do I want to take a gamble that we're gonna see a big surge after tomorrow? I mean, typically, I mean, if you look at these patterns, it seems like a lot of times, well, if you let's take NVIDIA for an example, a lot of times there's an earnings announcement and then NVIDIA's share price goes down. So is that thing, same thing gonna happen to to micron? I don't know. So I have to decide today whether I'm gonna pocket some some profits on Mic on on Micron. I mean, over the last I you know, I I've held Micron, I don't know, for several months, but if you look just at the last month of performance, it's up 40 percent over 22 days or whatever that is, right? So that's just insane. So I know it's come down a little bit and is basically flat today, right? Um as people try to make up their minds as to what to do.

Steve Davenport 19:45

Um yeah, I guess I don't necessarily want you to give me an answer. I just wonder if we could go through this just as an exercise for people I'm giving you my thought pattern.

Clem Miller 19:57

I'm giving you my thought pattern. Okay, the forward peg is really low, as I mentioned, 0.10. Uh what's the growth estimate? Um I don't have that here. Um wait a minute, growth estimate, long-term

Micron Earnings Scenarios And Next Steps

Clem Miller 20:10

growth estimate? Hold on. Oh, yeah.

Steve Davenport 20:11

What are you using in your peg?

Clem Miller 20:13

I got it, I got it, I got it. Hold on. Uh, where did I get that?

unknown 20:18

Hold on.

Clem Miller 20:22

Uh growth. Yeah, well, I'm you know, I'm taking a I'm taking from my source, I won't reveal my source, okay. Okay, okay, but , I'm getting 166 percent keger long-term growth rate, which is insane, right? How sustainable is that well you know in one evening we could see that 166 percent drop a lot, right? Uh so yeah, in other words, 0.104 peg could become a lot higher.

Steve Davenport 21:05

Yeah, I just wanted to say to people I don't I don't I don't believe 166. Yeah I'm gonna I'm gonna go out there and throw it into the ring and say I'm a I'm a skeptic there, come.

Clem Miller 21:19

Hard not to be, right? Well the other and and and as I pointed out, you know, I like to look at short at the short ratio, and this is this micron has the second highest short ratio of of any stock in my current portfolio. Uh and it's let me look it up here, and it's 3.33 percent. Now, in the grand scheme of things, 3.33 is not you know, it's not like GameStop, okay, right or AMC. So it's not super high, but it's Uh it's the second highest right now in my portfolio.

Steve Davenport 22:04

Uh and and you know I've got like I said usually if it goes over one or two, you look at that and say that's no.

Clem Miller 22:13

It used to be it, it used to be that if I if it went over 1.5 or especially over two, , I wouldn't want to look at it, right? But it seems like all of these I've loosened that up as all of these stocks have risen above 1.5 or 2. Okay.

Steve Davenport 22:34

So I but as the market goes up, you loosen your discipline.

Clem Miller 22:41

But I have my forward peg discipline, , which is not which as you've indicated is you don't apply any discipline to the growth rate. Right. Okay. Um if I if I applied a listen, if I applied a a short interest cap of two percent, or let's say 1.5 percent, I would have all cash okay in my portfolio. Okay. Um but you know, I'm offsetting that against the growth that we've seen, some of the some of the high sharp ratios we've seen. And so, you know, it gets me to questioning also conventionally speaking, a lot of people think that that short interest lower than five percent isn't bad.

Steve Davenport 23:32

Okay.

Clem Miller 23:32

I've always been on that more conservative than what most people think, right?

Steve Davenport 23:37

But you know, the yeah, Clem, you could be whatever you want, but if you keep changing the rules, you're gonna your discipline sound more like a wish than an actual community.

Clem Miller 23:50

The three point okay, look, my if I'm I'm just eyeballing this here, but it looks like my it looks like my average short interest median, median short interest ratio among the stocks in my portfolio is probably about 2.4, which which isn't isn't very high when you look at that. And and you got to keep in mind, Steve, is that this that's the the median among the stocks in my portfolio. And right now, and I don't have to worry, okay. So I don't have to worry as much about that because right now, believe it or not, I have 72% of my portfolio in cash right now. Wow. So I don't really have to worry that much about about short interest. I still worry about it, but you know, I guess if I had 95% of my portfolio in stocks, I would be very concerned about having a 2.4 median short ratio, right? Okay. So I I would I would really want to have a short ratio.

Steve Davenport 25:03

So you you try to you try to mitigate what's going on in an individual name on your overall portfolio risk. So when your overall portfolio lowers in risk, the names that you're willing to stick with might become riskier, but because they represent a smaller amount, you feel that's a reasonable approach.

Clem Miller 25:22

Correct. And this strategy, I got it, I mean, eyeball, you know, looking at it every day as I do, , not not making a trade every day, that for sure I don't, but looking at this every day, I tend to do better than the market on two-thirds to three market meaning S P 500 on two-thirds to three quarters of days, right? On the way down, , I do better. Okay. On the way up, I do slightly better. Um, I tend to do better on the way down when the market's down than I do when the market's up. Um but I still do better because of the of you know stocks like Micron and so on. And that may might change, right? And as I said, you know, if if if if I see a big surge in a stock in a particular day, then I I begin to question whether whether I should continue to hold that stock.

Steve Davenport 26:18

Yeah. I mean, I I think that it's a good discussion to have on the day before Micron reports. And I think that today, what I would say to our listeners is, and this is my summary statement, I think that Micron is going to be a leader in this space, and therefore it's also a much clearer way to get exposure to memory. Whereas SK Heinex and um Samsung are broad companies that have exposure to so when I want an isolated exposure and I want to be invested in memory, I think Micron might be the best play. And if there is an opportunity today on weakness, I think it might represent, you know, in my mind, the way I look at it, it might represent a good entry point for people who haven't had enough exposure to the AI trade. If you've already got this exposure, you've, you know, then you know you have to do what Clem's doing, which is think about whether you want to take a little off the table before they report tonight. If you think they're gonna be positive, I just look at the overall market sentiment right now, and it feels a little bit negative. So I have a feeling that it's gonna be more likely that they're gonna report good numbers, but their forward statement might be a little conservative because companies like to do that because you can't, it's easier to deliver on conservative estimates from management than it is on extreme estimates. So you can get a big hit on earnings day when you report beautiful forward forecasts, but then in reality, in three months you have to come back and back it up. And so, in my sense, I think it might be an opportunity after the bell today or tomorrow to think about you know putting it in a balanced portfolio of of exposures where you might be short in the AI space.

Clem Miller 28:21

Yeah, if if if you're starting off fresh.

Steve Davenport 28:25

How do you feel about micron earnings when you look at it tonight? I mean well summarizing.

Clem Miller 28:32

Well, I'm I I feel I think if I had to make a decision right at this moment, before the end of the day, I'll just stick with the micron exposure I have and see what happens, right? Because I've got so much in cash right now. Right. Um, I don't really have to worry about you know a big drop or not a big drop. I think beyond today, I think, yeah, micron's gonna do well, right? Um, but I think one has to be prepared for you know something of a of a shock tomorrow, um, negative shock, , should it happen. I and I would base that on what we've seen with NVIDIA. NVIDIA, despite all the hype, hasn't done very well recently.

Steve Davenport 29:14

Uh I mean, I think the number that you said earlier, the 40% in a month, probably should be a good indicator. I understand earnings have gotten better, and I understand forward earnings are tremendous, but I also understand that this demand cycle that we're in has to be put into perspective as to how long it can go and whether there's things that could disrupt. Um you have anything else you want to say about micron and memory?

Clem Miller 29:45

No, um , you know, other than I want to emphasize that NVIDIA parallel and um and say that you know it's quite possible that we could see micron share price flatten out um over the next few months, or you know, drop some tomorrow and then start you know bounce back a little bit and then flatten out. Um so but do I do I see it substantially and permanently dropping a lot? No, I don't see that. No.

Steve Davenport 30:18

So there could be an opportunity and maybe people in the right place, it could be an opportunity for people.

Clem Miller 30:25

Right. I'm just thinking, you know, do I want to take profits now or do I want to wait a year for them to to recover from a drop or six months from now to recover from a drop.

Steve Davenport 30:39

And and so that's a lot depends on whether it's in a taxable or a non-taxable account, right? Right. Yeah, exactly. Exactly. So okay. All right, yeah. Um well thanks for those insights, and I think that this did what I wanted to do, which is just talk about you know where is the micron earnings and how do they matter to the market. So thanks for everything,

Closing And What’s Next

Steve Davenport 31:05

Clem, and I appreciate your insights and about your portfolio. And listeners, um we've got two other podcasts we're working on. One on the memo the memo of understanding, which we think is gonna impact the market for the next three to six months, and also um SpaceX has it already had liftoff and is it going to go crashing back into the ocean? Um so we look forward to your listens and your comments. So please um share with your friends, and we appreciate you out there in Skeptics World.