저자: MeetKevin · 외국인
요약 보기
실업률 장기실업 지표 상승은 AI로 인한 구조적 일자리 감소와 부유층 소비 주도 성장이 원인일 수 있어 경기침체 신호로 단정하기 어렵다고 주장. 관세 수입 확대(연 2,550억달러 추정)가 경제에 끼치는 피해는 GDP 약 1% 수준으로 추정해 감내 가능하다고 봄. 엔비디아(PEG 기준 적정가 185달러)와 AMD(적정가 254달러)는 저평가로 판단, 테슬라는 PEG 4.5로 고평가로 판단. 엔비디아 실적 발표 직후 164달러 돌파 가능성 언급.
다듬은 글
So, I've decided to just lean into the whole orange thing and I'll make sure not to shave and then just keep running so I become more and more orange on a daily basis. I hope you enjoy the transformation over the summer and maybe by the end of the summer I'll actually be a full-time bull. That said, this morning in the alpha report, we speculated that market calm would continue mostly unlocking potential all-time highs for some momentum names like Cloud Coreweave uh and Tempest.
Although I was wrong about Cubits potentially coming back. Aqua was also on my list of four. So 75% hit on those that might momentum today. Uh and those that momentum momentum hard. But while I was looking at potential momentum movers, I was also looking at fundamentals because I like to look at those things separately, but sometimes there's overlap.
And so while I was looking at Coreeave, I went down a little bit of an Nvidia and AMD rabbit hole. and it led me to consider my fundamental bare thesis regarding something that seems relatively unrelated. But in this video, you're going to see how there could potentially be a very very clear link. See, uh, one of the reasons I've been a little bit of a poopy dupy bear, uh, is because since about the summer of last year, we've, uh, seen a a dramatic acceleration in the already accelerating path of the 27w week unemployed figure.
This is old news. We're not going to spend a lot of time on this. I I respect your time, so we're going to keep it fast. Okay. Basic bottom line is if you look throughout history, the darn thing basically never goes up unless there's a recession. See up recession? This time it started during a recession.
Sometimes it starts at the end of a recession. Sometimes it even goes up after the recession or it peaks after the recession. It sort of starts in the second half of a recession leading a lot of people to say, "Well, we're already in a recession and so this will all get graded in like a really long recession.
Uh, and the true peak of this unemployed period, long-term unemployed period will come in the next recession. Sort of like this right here, right? this 2019 rise and then oh it's COVID or over here you get a little bit of a rise in 2017 or sorry 2007 and boom oh it's the 2008 financial crisis right it's some form of crisis catalyst that ends up leading to some layoffs and this is typically true with the exception of very brief moments in time where you get these little mountains like 1995 over here you got a little bit of brief nothing like this uptrend we have over here or in 74 or 63 okay big deal so basically historically It would make sense if we went back and maybe painted honestly 22 on as some sort of longer extended rolling recession culminated by some massive shock at the end.
But is there a potential that this time is different? Understanding that the phrase this time is different is usually associated with no. Uh that said, is there a potential that this time could be different and there is a different explanation for why long-term unemployment would rise on on on sort of a structural basis.
And the answer to this is yes, there there is another reason for this. You this morning I was on a walk with Lauren and we were discussing how it's very likely that when Jack and Max, my seven and nineyear-old, enter the labor force, they're not going to be able to enter the labor force uh and and sort of work their way up anymore inside of any role that could really be replaced by artificial intelligence.
Now, that will be possible likely for a very long period of time because I think general purpose robotics doing general purpose service jobs, it's going to take an extremely long period of time, but think about it. You need people to bring your food to you. I mean, I know robots can do this, but generally like the food service at a table at the side of a busy, you know, street where you've got a restaurant, somebody to pull your waverunner out for you, a lifeguard, uh, you know, lifeguard augmented with AI.
Now, there's an idea. Who's got an app for that? you know, Palanteer is gonna beat that one. Uh, but the point is, uh, construction jobs, right? Anything anything physical is probably the last frontier where things really get, uh, innovated away. Uh, but what we thought is what you'd probably end up seeing, and we've talked about this on the channel as well over the last few years, is really a rise in the long-term rate of unemployed because as people lose their jobs, they just can't get jobs again because they're being replaced by artificial intelligence.
Uh, and and then those replacement jobs just aren't available at other companies because other companies are like, "Well, yeah, I don't I don't need your skill. You've already been replaced." And this is where we see these articles like here's one. IBM slashes around 8,000 jobs primarily from its HR division thanks to artificial intelligence.
You know, this is just a recent piece here, but we get posts like this on almost a daily basis. Uh, I mean, when I got back, I even saw a, you know, a cover of, um, an Atlantic story, and I know people like bagging on the mainstream media, and that's fine, but, uh, you know, AI is breaking entrylevel jobs that Gen Z workers need to launch their careers, LinkedIn executive warrants.
Yeah, fair. Because that's sort of where you sort of gain your experience and then you could go on and actually be an executive somewhere else. Like somebody doesn't get a job as a CEO off the street with no experience, but if they've had previous executive level seuite experience, it's a lot easier for them to get in.
Well, how do you get that first set of experience? Well, often you work your way up at a company. But if those opportunities get replaced by AI, somehow you have to either know somebody to get in or you got to figure something else out or you know, and while you're trying to figure something else out, where does, you know, where do you end up?
Well, you end up potentially in this figure right here, this longerterm unemployed figure. So, is it possible that this chart has actually just been rising since the advent of frankly GPT? Uh, and the answer is yes. In fact, if you pin this over here, it represents uh roughly the beginning of 2023 when we started seeing the longerterm trend of unemployment rise.
Now, this is actually really interesting for uh for the bears out there, uh you know, speaking also to myself uh when it comes to this unemployment chart because I I see it as as a warning this chart. It's this is not good. Historically, this is a very bad chart. But again, could this time be different uh this data point?
Well, yes, because it is possible and this is kind of weird, but it's definitely possible. It is possible that growth is being driven uh by not just artificial intelligence but by the wealthier cohort of spenders or corporations. So wealthy companies think you know Fortune 500, MAG7 whatever uh wealthier sort of 1% individuals they spend more and let's say their contribution to GDP is positive 2%.
And poorer individuals uh are spending you know on net 1% less towards GDP. Well, so now let's say if GDP is going to be 2%. But you get a 2% boost from all the rich spending and then a minus 1% drag, you're still at 3%. Or, you know, you could also make the argument that maybe we're only at 1%. Minus one because of the poor spending in the toilet and delinquencies rising and pain at poor levels, but then the wealthier cohort adding two back in and oh wow, you're at 2.2% which is exactly where the Atlanta Fed real GDP estimate is now.
So really AI is kind of the perfect boogeyman explanation for this. So in other words, the rise in the longer term unemployed chart may be a symptom of growth rather than weakness, which would be the exact opposite of what has traditionally been true. Crazy. Something that would otherwise be an indicator of problems could actually just be a symptom of the growth that we're seeing in our economy.
scary for those who are losing their jobs, scary for bears relying on this indicator, but also kind of good for America broadly. Anyway, speaking of America broadly, then there's also the question of how much damage tariffs are doing. Uh there is uh this talk right now that if you analyze how much tariff revenue we are making, we are making about $255 billion in tariff revenue.
The question though is how much more how much greater is that from normal and what kind of damage is that going to cause to the economy? Well, I'm about to give you some examples of exactly what kind of damage that could cause. But first, a message from our sponsor Lear Capital on diversifying. It's official.
Meet Kevin has finally partnered with a gold company. But . And that's what Lea Kevin sent you. Okay, so here's the scope. Look, previously we collected about $85 billion from tariffs. Now, we're collecting on an annual runway 200 run rate $255 billion. That could obviously go down if trade shrinks because of that.
But let's just say it doesn't and we collect 174 million a billion dollars more in tariffs uh in tax revenue every single year. Now let's say the damage of that $170 billion in tariffs is twice as impactful. So we are hurting our economy by $340 billion. Okay, that's about 1% of GDP. So once again, if we're growing at 3% and we damage our economy with 1% tariffs, then we're at 2%.
or if we're at 22, which is where the Atlanta Fed thinks we are now, then we're at one two. Big deal. Still positive. Again, potentially because of the AI factors, but also the fact that, you know, tariffs just aren't that big of a deal at 10 to 18% or whatever. So, we'll see where they sit now. As long as they don't get substantially worse, could actually be quite solvable and and investable, maybe.
Which that then begs the question of all right then, are Nvidia and AMD a screaming deal right now? And could one of these double or potentially both of them? And that's actually an important segment in what we've got to talk about here because yes, while I'm a big fan of f you want to invest in artificial intelligence, these might be some opportunities.
So, we'll talk about the numbers on these. And when I talk about diversifying away from AI, you got to think what kind of jobs are still going to maintain. Well, we talked about this physical uh labor style service jobs at least until robots take over the world. And I also think real estate and development, fixing up a fixer upper, you know, we find it, we use AI to to find deals, but like actually fixing it up, hiring people, contractors, coordinating work, and then of course actually developing properties from scratch.
All of these things matter. They're things that we have to work with. So this leaves me with an analysis on Nvidia and AMD to consider. So Nvidia and AMD uh recently have been selling for in my opinion below where they they should be selling for. There are probably some reasons for that. I think there is fear in Wall Street that their growth rates are going to be revised down and that the stocks are already discounting uh a lot of the value of these revisions down uh either because of uh tariffs or China or you know a slowdown in AI or whatever.
But what we really have to understand when it comes to these companies is where they're valued now uh for options purposes. Their volatility is super low right now. But when it comes to their valuations, I like to use a PEG ratio based analysis. Now, some people use this in a very different way. So, I'm going to explain the way I do it versus what other people do.
Other people will say, "Well, Kevin, Nvidia has a PEG ratio of under one right now, which suggests it's a screaming buy." Okay, but does it? Well, let's analyze it. Well, if you take Nvidia's price, 135 bucks, divide their earnings per share for January 2026 into it. I don't like using the trailing 12 months.
Divide that in, what do you get? 30.9. Okay. What is their next four years of growth expected to be? 15.9. So, PE ratio 30.9 divided by 15.9, what do you get? 1.94. That's the PEG ratio that I think Nvidia sits at now. Now, I personally think their growth rate will be closer to maybe 12 or 10% over the next four years, but that's fine.
If you compare them obviously to the growth rate that's already happened in the past, which Wall Street doesn't care about, then yes, you will get a PEG ratio under one. So, I always like to use a consistent PEG ratio, look at the end of the year's earnings per share ahead of us, and then four years of growth thereafter.
That's how I get my PEG ratio. So, in my opinion, for a service business at a 26-9, sort of a justifiable PEG ratio, Nvidia could be $185 company and AMD could be a 254 uh well, sorry, let me make sure I said that correctly. Nvidia could be a $185 stock. Okay, now talking about the market cap, what the stock price could be $185 and AMD could be $254 because their PEG ratio is only 1.21.
And I think both of them belong somewhere around 2.6. 69 because that's generally where I price software companies. And when I do my PEG ratios like this, I'm able to consistently compare across different industries uh and various different companies. For example, when I look at Tesla, I think that Tesla's a little bit pricey right now.
Mostly because if I look at Tesla, first of all, they're a manufacturer. Like they actually manufacture their own products, whereas Nvidia and AMD do not. you know AMD for example doubling down on uh UA link in their consortium partnership with Dell and others to compete against Nvidia's Infinavan uh these are not products they're manufacturing they're designing them into the uh you know infrastructure for these companies uh for the data centers mostly and there's all this fear about walled gardens like oh Nvidia is going to wald garden us like Apple so we need to diversify our technology whatever it just bodess well for AMD but I think the whole industry grows And people are going to take what they can get for a meaningful time in the future until AI hits a wall.
But look at Tesla for example. It closed today at 360. It's actually impressive that it broke out over the U 347 level. And uh the earnings per share that are projected for Tesla, they're actually pretty darn good right now because earnings are expected to be so bad this year at just 193. So take 360 divided by 1.93 for the year.
Tesla's technically right now trading for a 186 PE ratio, but Wall Street broadly thinks they're going to return to substantial growth next year at 50% growth, 32% growth, 33% growth, and 49.7% growth. These are some of the highest expectations that I've seen for Tesla's growth in a very long period of time.
This has always averaged like 28 to 32%. And now if I divide this by four, I get 41% growth. That's a lot. The problem is I have to now divide it by the price, right? The PE ratio. So 186 divided by 41. This sucker's trading for a 4.5 peg. That's insane. So if I do 41 times growth and I take uh 193 on EPS and I do a 167 peg ratio, you know, we should be at about a buck 32.
So $132 per share on a simply auto manufacturer and energy business EPS POV including EPS growth rates in the future. So already including the next four years of growth rates assumed by Wall Street in earnings per share which does discount some earnings for Optimus, robo taxis, semi-truckss and all of that.
So in other words, it's extremely expensive. And when I compare the three companies, this is a tool that I like doing. Now, don't get me wrong, I still like Tesla. I still like my Cybert truck. Uh, you know, I I just think the valuation is a little kooky right now and full of a little bit of opium, which is great because it makes people a lot of money.
And I think it's a great opportunity to sort of diversify or set some trailing stops, right? Like milk it while it's going up and then stop out 20% on your way out so you build up some cash. Like what's wrong with that? So anyway, uh that said, Nvidia obviously has earnings tomorrow. Uh oh, what bottom line of that is that sort of lets me adjust what I think things are worth.
But if we look at Nvidia's earnings tomorrow, you know, we're expecting earnings per share of 87 uh uh per share. So call it 88 cents of adjusted EPS. Uh non-adjusted, we're at about 79 cents, revenue 43.3 billion, net income of 21.6. Nvidia typically beats. I mean, we're talking seven out of eight to eight out of eight times.
So basically always they always beat. It's just a matter of how much and then what they say about Blackwell. I mean, honestly, Nvidia's been on a little bit of a of a downtrend for a while. Maybe these earnings will finally be the earnings that let them break out. And if they break out, they'll break out to 164.
Take a look at this. Look at this ceiling you've got right here. Reject basically, you know, this downtrend kind of continues. You know, you could draw this probably a little bit more cleanly, but whatever. You get the idea here. This is sort of the channel. That's what I drew is the sort of channel that you're in right here.
And I think you could break out of this and get to 164, which is uh that's a fib extension. So yeah, that could be very interesting for Nvidia. 164 would could be a reasonable target afterwards, you know, unless they have some kind of like bearish topic regarding growth or whatever. But I don't know.
I I guess maybe I'm just not seeing that bearishness. I do see competition from, you know, AMD. Uh and you could directly see that competition because some data centers are now concerned that Nvidia is going to sort of walled garden them in like Apple does. That's actually led to a rise in AMD's UA link.
Uh this is a consortium by like Dell and AMD and others to basically compete with Nvidia's Infiniband. Basically wiring your data center with Nvidia components and switches and everything. So like everything is married together and really happy and it's all the same product, the same company. It's all unified.
Uh kind of like ubiquity. Uh well, you know, there have been some walled garden concerns and that's what's given AMD a little bit of a rise here. And so, you know, AMD's been ramping up their MI 350. Uh you know, that's going to be especially important here now because of all the Chinese comp, you know, complaints about, oh no, we're not going to be able to sell some of our Chinese products to China.
Yeah, whatever. Uh like they'll be fine. So, I think right now the only thing that's kind of keeping AMD and Nvidia stock growth down, like how much they're returning year to date, is fear that their growth will be revised down. I think the stocks are kind of already discounting that off the pegs because like I don't think I said it, but you know, at 169 justifiable for Nvidia, you're at like 185 current Wall Street estimates, right?
Mine a little lower. Uh$ 269 for AMD, you can be a $254 company or stock, right? So like that's a double. That's nice. So it it's looking good. You know, I think some of the China tariff fears have already been priced in. So basically, you have these two massive price or cash cows. The fears about discounted growth and tariffs, I think, are kind of already built into the stock.
And I mean, it would really have to take hitting like a wall or something to get to get, in my opinion, these these stocks to just perform really poorly over the next few years. But it's possible that they do, you know, should that wall hit. I I mean I guess we'll get a little bit more color tomorrow, but I mean I'm wearing my Jensen jacket because Jensen is not the guy who's going to tell you bad news.
That said, that doesn't mean the stock's going to go up. Last time Nvidia had earnings was February 26th and the stock went straight down right after that. Now, in fairness, that was also in the leadup to Liberation Day. So, a lot of things went down over that period of time. We were in a little bit of a bearish downtrend. So, who knows? Anyway, that's what I got for y'all tod.
Comments 0