Bloom Energy拥有强大的技术和改善的利润率，但对Oracle的严重依赖和其他集中风险使其成为短期交易而非长期基本面买入。

Nancy Pelosi's periodic transaction report, which is a fancy way of saying she bought the dip and it was probably her husband, but it doesn't matter because dang, they just spent millions of dollars buying Bloom Energy and Intel.

I see at least $15,000 worth of share purchases. I see call options that were purchased with a strike price of $100, so well in the money for a year out. They made these purchases on July 24th, so as usual, it's been 2 months. Oh, actually no, it's been exactly 30 days. It's been exactly 30 days since the transaction, so of course we get the news late, but if you keep in mind what was going on a month ago, you'll remember that that was exactly when Leopold was getting Leo folded. Who knows what kind of insider informa- info Nancy Pelosi had, but that's okay because on the 24th, the stock traded between $183 and a high of $218. That's literally in the strike range for where Bloom Energy is trading now at $205.

So, is it a good deal and should you buy the stock? What even the heck do they do? Okay, let's keep this really simple because obviously it is well down from its high of $351. That is a fall of 61% from peak in the value of this company. So, what drove that? Fundamentals? Momentum? The answer is both.

So, here's the thing about Bloom Energy. Bloom Energy is a fuel cell provider that gets fueled with natural gas and then chemically converts it to energy, so it's a little bit different than just like lighting it on fire, like a you know, a gas engine for example.

Uh but what's fascinating about Bloom Energy is they've had really bad margins and because of this data center boom, their margins are blowing up. And when you look at their income statement, it's really good right now. Now, it didn't used to be. They had negative margins in '21, '22, '23, '24, '25. You get the idea. But now, look at this. Their revenue has 2.65x'd to over a billion dollars. Their gross profit is sitting at about 33 to 35-ish percent. They're They've literally got gross margins on their product that are like 37-ish percent, and then of course you've got like install and some of these other lower margin things. Uh they bring about 18.5% to the bottom line now, and in their earnings call they're guiding about 21% that they can bring to the bottom line, and about 34% that they can bring to the top.

Now, why does 34% on gross margin matter for the full year? Well, because that's really similar to Apple, which is kind of exciting because, you know, Apple's a pretty high margin business. We know their stuff isn't exactly the cheap for a pretty penny. But if you go to any Apple financial statements, you'll see that their products also sell for about a 37% gross margin. Now, the thing that's different about Apple that gives Apple a big premium is they sell services at about a 70 to 75% gross margin that Bloom Energy doesn't. So, the Apple products that are selling for about 37% margin, so you know, you sell $100 worth of stuff, $37 goes into the business that they can then spend on research and G&A or advertising or whatever. Then they pay taxes on it, then they get their net, right? They got $37 $37 out of a hundred left. That is great because it doesn't just give you the $37. It leads more people to sign up for subscriptions to iCloud or, you know, whatever the the life software that they have for music or the subscriptions for any of their other products. Or just included. That's a big W for Apple because it's really high margin. Bloom, on the other hand, you have to think of it as just mostly, the vast majority of it being the higher gross margin manufacturing side. The reason that's important is because it caps their net. We're really not expecting to see the net income for Bloom to really ever go beyond 25% because they don't have that services side. They're still making things that cost a lot of money.

And these fuel cells are great because you can stand them up in three to four months, whereas, you know, a natural gas facility at a data center might take you somewhere around three to four years to stand up. So, there's been this huge boom in let's just use fuel cells to stand up our data centers so we can get up and running and lower our time to revenue on data centers, which is really important because guess the biggest risk for Bloom Energy? Their main customer. So, this is customer concentration risk, which I think is a really big red flag. That's not to say I'm bearish on the company. We'll talk about a price target in just a moment, but they have got serious revenue concentration. 44% of their Q1 or their last quarter's revenue came from Oracle. And in the last six months, 73% came from Oracle.

Well, who's picking up that debt? It's Bloom Energy, which is now getting a big momentum boost because of Nancy Pelosi. Intraday, it's up like 11% just intraday on this Nancy Pelosi disclosure riding right here at my 206 line, which is a fib retracement line. Uh and maybe we'll be able to sustain that and break back up, but you could see since uh June, after the Google and SpaceX suckening, the stock has been falling. Why? It could be visibility into growth. And that's always what's really important about these companies is we try to forecast growth.

See, Apple's growth isn't that high, but it's stable. Apple's growth is like 9% stable. And that means they're trading for about a 3.7 PEG right now. In my opinion, Apple's actually about 40% overvalued right now because I like to combine, you know, a 35 PE ratio divided by the growth and go I don't want to pay that much for that growth. Now, there's hope that the CEO is going to help transition the company to more products and more AI revenues, but a lot of that hope is already priced in, all right? Bloom Energy on the other hand, let's look at the forecast for Bloom Energy.

So, if I want to go really aggressive on the upside and I give them a really juicy like closer to software level price to earnings growth level, I could take their forecast earnings per share of $2.55, give it a multiple of 2.4 * 50% growth, I get a 306 price target. But the problem is the growth that's going up to 50% is going to taper off, especially as Oracle buys fewer of these in the longer term. It's expected in 4 years, 3 and 4 years to taper down to 25%. And so if I run this at just 25% growth times a 2.4 multiple times their earnings per share today, my fair value today is actually 153. So at 205 you know, we're we're we're closer to this, but like what's my upside, right? My upside is expecting that this can keep booming and growing forever, but I'm really reliant on Oracle. So what that means is I need to expand to different customers. I need to sell to Oracle more. I don't want to bet on Oracle. Uh or I need my margins to go up. Margins they're good but we're already seeing some of the pressure here, right? Revenue's up 2.6 times and product costs are up 2.9x. So costs are up higher than revenue and they say part of that is because they had to pay an increase in stock-based compensation on their sales, which to me suggests that like their sales people are grinding really hard to get those sales and contracts signed. And if I got to pay more than I'm able to increase the prices or revenue, then it just sort of makes me question like how long and sustainable is that revenue?

So I'm not saying this is a bad deal. I'm saying there's concentration risk. Uh there's a lot of momentum in it and the company is growing margins. I got to give them that. They do acknowledge that margin swings a lot. They say it over here. Hey, we've been negative. We went public 8 years ago. In between that time we've swung from negative to positive. There's so much going on uh in terms of changing, but we're really bullish on our total addressable market. We think there are massive tailwinds. We're going to be able to increase market share, get more customers. You know, maybe Oracle is just the start. They also have a related party that buys their own products. This is a fund that they basically operate in partnership with Brookfield Asset Management. And so in some of their concentration risk, you'll actually see, "Hey, some of our sales went to a related party entity." That's their AI fund. And their AI fund naturally buys Bloom Energy products. But again, in the last quarter, that represented 21% of their revenue. So their own sort of circular financing was 21% of the revenue. Oracle was 44% in the last quarter. The quarter before that, Oracle was even more, right? So if you lose the circular part, if you lose Oracle, you lose a lot of that growth momentum that we're seeing. Now, hopefully you could just land new customers, but that's the bet that you're making on this income statement.

Now, cash flow looks good. Cash flow yield a little low, though. Okay? We've got about a 1% less than 1% cash flow yield. Uh you've got Palantir is also about a 1% cash flow yield. Very different from a software stock that is going to be trading at uh you know, like a 10% cash flow yield. An example of which might be like Salesforce. Salesforce was one of the buy alerts that we sent at about $163 per share.

So uh let's just continue and uh look at a little bit of their balance sheet. So their balance sheet is decent. We've got $2.6 billion in cash. And I've got $800 million in bills to pay. So I'm pretty decent on the balance sheet. I do have 2.4 in long term. Though I will say those long terms are mostly uh convertible liabilities for 2030 that have a 0% coupon. So, it's not like they're dying or drowning in debt here when they got a 0% loan, basically. It's pretty good.

So, there's a lot to be excited about when it comes to Bloom Energy. And I get it. I get the momentum. I think the technology is brilliant. I think it's really, really smart. Uh but, I do think there are some risk factors. And so, our stock AI tool, uh which basically tries to use my brain and tune algorithms in combination with artificial intelligence, gives you a couple forecasts. We'll look at both Apple and we'll look at Bloom Energy.

If I look at Bloom Energy, I see it at 124 with this target on stock AI. Now, why? The reason we sit lower there is because the examples I gave you on the income statement were the more aggressive style PEGs. I think 306 to 153 requires a higher net margin that this company is not going to achieve. So, more realistically, you're probably somewhere between, depending on what growth ratio you use between 90 bucks to 178, which means for me, this is more of a short-term trade than it is a fundamental hedge buy at this point. For a bottom-line takeaway on Bloom, the big play here is can they get more customers and can they grow fast their margins faster? That's the real play on Bloom. My take uh is that you really want to see expansion not only to customers, but margins both happening. What that's probably going to rely on is not just an expansion of what artificial intelligence is used for, whether it's in healthcare, robotics, you know, space, you know, whatever AGI, which, you know, I don't believe in AGI, but whatever. Um what we really want to see is an expansion of that boom, kind of like what we talked about in that memory cycle, where you can expand out that data center plateau before you kind of get that eventual decline. Ideally, you get some more S curves of AI data center growth and you really get fuel cells becoming sort of a core product for not just Oracle uh that they purchased, but also other data centers. So, that doesn't make Bloom Energy a bad company. It's a concentrated risk. Now, that doesn't mean companies with concentrated risk are bad. Like it's just who's the counterparty. So, in this case, the counterparty is Oracle, which credit default swaps are skyrocketing on. We really want to see expansion to other data centers rather than just Oracle.

A company like Cerebras has a lot of customer concentration in uh you know, ultra-fast token artificial intelligence, ultra-fast token generation. Nvidia competes with their Groq product, which they just announced is in uh full production. Uh but this is really useful for customer service and voice artificial intelligence, low-latency communication. But Cerebrus has really big customer concentration with OpenAI. You know, OpenAI is going to be like, you know, an Oracle is for Bloom Energy. That needs to expand. And if Cerebrus can't expand past OpenAI, that's bad. And so, you know, then you kind of get stuck and people just use the Nvidia product and the Cerebrus product doesn't go anywhere, especially if Nvidia can scale faster. Those are all real risks.

So, it just depends on kind of where you want to place that bet. And then, as long as you place that bet and know that, then you can kind of monitor how things are evolving. Is margin increasing uh and is that customer concentration uh declining toward a broadening out? Uh and then, of course, it would help if the entire artificial intelligence boom kind of keeps going to expand that plateau. That's pretty uh critical for uh Bloom Energy at sort of bottom line on it.

So, there's been this huge boom in let's just use fuel cells to stand up our data centers so we can get up and running and lower our time to revenue on data centers, which is really important because guess the biggest risk for Bloom Energy? Their main customer.

So, this is customer concentration risk, which I think is a really big red flag. That's not to say I'm bearish on the company. We'll talk about a price target in just a moment, but they have got serious revenue concentration. 44% of their Q1 or their last quarter's revenue came from Oracle. And in the last six months, 73% came from Oracle.

Uh and maybe we'll be able to sustain that and break back up, but you could see since uh June, after the Google and SpaceX suckening, the stock has been falling. Why? It could be visibility into growth. And that's always what's really important about these companies is we try to forecast growth.

But the problem is the growth that's going up to 50% is going to taper off, especially as Oracle buys fewer of these in the longer term. It's expected in 4 years, 3 and 4 years to taper down to 25%.

And so if I run this at just 25% growth times a 2.4 multiple times their earnings per share today, my fair value today is actually 153. So at 205 you know, we're we're we're closer to this, but like what's my upside, right? My upside is expecting that this can keep booming and growing forever, but I'm really reliant on Oracle. So what that means is I need to expand to different customers. I need to sell to Oracle more. I don't want to bet on Oracle. Uh or I need my margins to go up.

Margins they're good but we're already seeing some of the pressure here, right? Revenue's up 2.6 times and product costs are up 2.9x. So costs are up higher than revenue and they say part of that is because they had to pay an increase in stock-based compensation on their sales, which to me suggests that like their sales people are grinding really hard to get those sales and contracts signed. And if I got to pay more than I'm able to increase the prices or revenue, then it just sort of makes me question like how long and sustainable is that revenue?

They do acknowledge that margin swings a lot. They say it over here. Hey, we've been negative. We went public 8 years ago. In between that time we've swung from negative to positive. There's so much going on uh in terms of changing, but we're really bullish on our total addressable market. We think there are massive tailwinds. We're going to be able to increase market share, get more customers. You know, maybe Oracle is just the start.

They also have a related party that buys their own products. This is a fund that they basically operate in partnership with Brookfield Asset Management. And so in some of their concentration risk, you'll actually see, "Hey, some of our sales went to a related party entity." That's their AI fund. And their AI fund naturally buys Bloom Energy products. But again, in the last quarter, that represented 21% of their revenue. So their own sort of circular financing was 21% of the revenue. Oracle was 44% in the last quarter. The quarter before that, Oracle was even more, right? So if you lose the circular part, if you lose Oracle, you lose a lot of that growth momentum that we're seeing. Now, hopefully you could just land new customers, but that's the bet that you're making on this income statement.

And so, our stock AI tool, uh which basically tries to use my brain and tune algorithms in combination with artificial intelligence, gives you a couple forecasts. We'll look at both Apple and we'll look at Bloom Energy.

For a bottom-line takeaway on Bloom, the big play here is can they get more customers and can they grow fast their margins faster? That's the real play on Bloom. My take uh is that you really want to see expansion not only to customers, but margins both happening.

What that's probably going to rely on is not just an expansion of what artificial intelligence is used for, whether it's in healthcare, robotics, you know, space, you know, whatever AGI, which, you know, I don't believe in AGI, but whatever. Um what we really want to see is an expansion of that boom, kind of like what we talked about in that memory cycle, where you can expand out that data center plateau before you kind of get that eventual decline. Ideally, you get some more S curves of AI data center growth and you really get fuel cells becoming sort of a core product for not just Oracle uh that they purchased, but also other data centers.

So, that doesn't make Bloom Energy a bad company. It's a concentrated risk. Now, that doesn't mean companies with concentrated risk are bad. Like it's just who's the counterparty. So, in this case, the counterparty is Oracle, which credit default swaps are skyrocketing on. We really want to see expansion to other data centers rather than just Oracle.

But Cerebrus has really big customer concentration with OpenAI. You know, OpenAI is going to be like, you know, an Oracle is for Bloom Energy. That needs to expand. And if Cerebrus can't expand past OpenAI, that's bad. And so, you know, then you kind of get stuck and people just use the Nvidia product and the Cerebrus product doesn't go anywhere, especially if Nvidia can scale faster. Those are all real risks.