When you don't know what to pay for a stock you tend to sell it. Right now, we have people selling technology shares left and right because they are unsure of what they own and what it might be worth. They are looking at the incredible shrinking price-to-earnings ratio — the "multiple" — and they are saying get me out of here. I see it. And not just in tech. What do we pay for Danaher when it buys a company , Masimo, that we have long since disliked, ever since its litigation with Apple over patent infringements? You don't sue Apple and make any friends. Plus, Danaher went downscale into Masimo's pulse oximetry market. At just under 25 times next year's earnings, I have to say no thank you. This is a company that has not "earned" its multiple in many, many years. Let's also look at Workday , a software-as-a-service (SaaS) company where you pay by the seat for their HR and financial software. With those characteristics, you have to wonder if 15 times next year's earnings is the "real multiple," meaning that maybe next year's earnings estimates are way too high. When co-founder Aneel Bhusri previously had the CEO title, he was "money" for investors. But I was not impressed with his successor, Carl Eschenbach, who stepped down Feb. 9 to be replaced by none other than Bhusri. I caught a shiver down my spine last month when Eschenbach said the reasons why Workday won't be disrupted by artificial intelligence include the relationships the company has established. Some would say Bhusri is Workday when it comes to the big relationships. What happens now? I can't believe that multiple will turn out to be that low. Some investors would much rather be in Micron at roughly 10.5 times next year's earnings with the chance that the multiple could expand because of how proprietary Micron's high-bandwidth memory (HBM) chips are. They are in short supply and coveted for AI computing. In fact, the multiples for most of the memory and storage stocks are low, amid a serious shortage that is leading to profit windfalls for them. Sandisk is at 9 times earnings, even after an incredible 1,219% rally over the past six months. You think you missed the move, but others are getting into the stock on hopes that you could get an upside earnings surprise and a higher P/E ratio on that surprise — the best kind of stock rallies, as I describe in "How to Make Money in Any Market." With a P/E ratio of 24, Seagate's multiple is a bit richer than Micron and Sandisk. But that's still only marginally above the broader market multiple of 22 while the company is projected to grow earnings by more than 50% this fiscal year. Western Digital is slightly richer than Seagate, selling at just under 26 times forward earnings. In this historically cyclical industry, I would prefer those with the cheapest multiples. In any case, my point is that the disparity in valuation in this cohort demonstrates how hard it is to figure out what to pay for stocks in this market. More industries, more problems There are valuation problems in all sorts of industries, not just tech. Club name Capital One may turn out to be the fastest grower in the banking sector, but President Donald Trump's threat to cap credit card interest rates will likely make it difficult for the stock to fetch anything above a 10 multiple. Plus, Capital One's pending acquisition of expense management provider Brex — seen by some as inferior to fintech rival Ramp — adds more execution risk on top of the Discover integration . Hence the pathetic multiple, down from 13 times earnings a year ago. Capital One now has a lot to prove. I think it can do just that, but it needs help from Congress to ensure there's no cap on rates. For years, the stock of Goldman Sachs traded at a discount to JPMorgan Chase , in large part because of its episodic earnings tied to investment banking cycles. But, as former Goldman chief Lloyd Blankfein writes in his fabulous new book "Streetwise," CEO David Solomon has been able to smooth out its earnings. And now Goldman trades at a P/E above JPMorgan — currently at 15.8 times forward earnings versus 14.4 for Jamie Dimon's behemoth. It was Solomon's mission upon taking over in 2018, and it's why he is such a good CEO. The most mysterious valuation reassessment — or one might say "guessing" — is still in technology, though, and it's gotten quite painful. Ahead of next Friday's Monthly Meeting for Club members, let me flesh out the irrationality weighing on about 40% of the market. CRWD .NDX 6M mountain CrowdStrike's stock versus the tech-heavy Nasdaq 100 over the past six months. We'll start with how news is handled when your stock is out of favor. CrowdStrike and Palo Alto Networks are two stocks that we've made a great deal of money owning, but we're now risking a big giveback — something I believe is unforgiveable. CrowdStrike announced exceptional news Wednesday: Its ecosystem of cybersecurity tools, called the Falcon Platform, is now available on the Microsoft Marketplace , which is essentially an app store for Azure cloud users. Do you know how big this could be? CrowdStrike does $1 billion a year with Amazon's equivalent service for Amazon Web Services. The company, which is projected to do about $4.8 billion in revenue this fiscal year, does zero with the Microsoft Marketplace. This could be a huge deal for CrowdStrike, especially because customers can use their Microsoft Azure credits to buy Falcon protection. CrowdStrike is about 27% below its 52-week high. Still, this news meant nothing to this company's stock Wednesday because it is very expensive at 85 times current earnings. The fact it landed on the same day that fellow Club name Palo Alto Networks dropped almost 7% on disappointing earnings guidance didn't help, either. For its part, Palo Alto's stock sells at 39 times earnings estimates for the next 12 months. That is down a great deal from where it used to trade, commanding a 55 multiple as recently as the fall. The stock price has also retreated significantly, going from $220 in October to the low $150s now. Again, this one is hard to value. CEO Nikesh Arora proclaimed foul when it came to comparing its guidance to Wall Street's consensus estimates, saying there's noise around its CyberArk and Chronosphere acquisitions that distorts the evaluation. Perhaps more importantly, the market is worried that Palo Alto will be crushed by AI, lumping cybersecurity into the broader SaaS bucket. I have trouble with this. I don't think companies can use AI to replace the comprehensive security that Palo Alto provides. If you ran a company, do you really want to let a large language model be your protector? It doesn't matter, though, the community has spoken and voted Palo Alto off the island. Giant conundrums Tech giants Alphabet , Meta Platforms , Microsoft and Amazon aren't immune from quandary status. We have been slowly but surely buying back Alphabet because it sells at just 26 times forward earnings, despite having YouTube, Google Search, the Gemini AI chatbot, the Chrome browser, driverless taxi service Waymo, and the fast-growing Google Cloud service. That's way too valuable an agglomeration to sell at 26 times earnings. Plus, the stock has pulled back more than 11% since Feb. 2, so you may be able to get it for even less. I don't think it will stay down here for long, though. Not with that lineup. We bought more on Tuesday. Meta seems astonishingly cheap at 21 times earnings. But right now, it is valued as an advertising play and you are not going to get a higher multiple than it has from the ad business. We learned Tuesday night that Meta was committed to buying millions of Nvidia chips . Why would it do that? Why would it spend $6 billion locking up fiber-optic data center cabling from fellow Club name Corning ? I postulated Wednesday that perhaps Meta wants to develop a cloud-computing service of its own. That would be terrific for Nvidia but a disaster for Meta's balance sheet because it isn't cheap. Maybe Meta committed to buying all that Nvidia equipment to get bragging rights over Google and Amazon, which spend a lot of money on in-house chips that aren't as good as Nvidia? I don't know, hence the challenge to figuring out what to pay for Meta's stock. Microsoft at 22 times earnings — essentially on par with a market multiple — seems like a steal. But we must still deal with that last quarter and the unfathomably bad Microsoft 365 Copilot product, which has 15 million paid subscribers out of over 450 million M365 commercial seats. And yet, management bragged about that 3% conversation rate on the earnings call. Terrible. Why not sell it? Because of optionality. The moment we sell may be the moment that the company bites the bullet and creates a new product that is the envy of the entire industry. When that happens, we will look back and say, "What were we thinking? They had so much optionality and you sold it because of Copilot?" I will not make the same mistake twice — I did that with Google when we initially exited last year. Any company with this much money can and will do something that generates a better return than Copilot. I do not want to sell Microsoft at the bottom. AMZN 6M mountain Amazon's stock over the past six months. Finally, there is the most problematic of all: Amazon. The stock had a nice reversal Tuesday, followed by another move up on Wednesday — its first back-to-back winning days since Jan. 23. Indeed, it's coming off a nine-day losing streak that wiped out more than $450 billion in market cap on the heels of its $200 billion capital expenditure plan . They're spending all that money to make Amazon the best in the AI service world. Somehow, though, the market has spoken and said that title belongs to Google. I don't believe this is a settled debate. I think Amazon can unlock an amazing amount of value if it so chooses. Part of that is being able to build data centers faster to convert them into revenue-generating AI factories. Tall order. That brings us back to the question: What is Amazon's stock worth? Can we give something that misses earnings estimates for the quarter a 26 multiple? Absolutely not. It's too high. I can't think of what Amazon can demonstrate between now and its next earnings report that can change its stock's fortunes. So, the short-term forecast here is pain because Amazon is obliterating its cash flow with all of this capex. The SaaS collapse Workday can't be the only SaaS name we cover in detail. Let's zoom in on ServiceNow , though we don't own it for the Club. I have to tell you that I have never seen a stock more staunchly defended by management than ServiceNow. Alongside its earnings report in late January , it announced an additional $5 billion for its buyback program, including a $2 billion accelerated share repurchase to get in there and start buying quickly. And on Tuesday, a number of executives — including CEO Bill McDermott and Gina Mastantuono, its president and finance chief — canceled their preapproved stock sale programs. McDermott also signaled intentions to buy $3 million worth of stock next week, the earliest possible date he can do so without trigging a regulatory rule that requires him to return profits from recent sales. That is an extraordinary stand, and yet it's meant nothing. The stock fell 1% Tuesday despite all that news (though it did add 1.8% in Wednesday's session). ServiceNow has gone from $129.62 before earnings to roughly $108 a share now. This fast-growing company with a "Rule of 56" — calculated by adding together its subscription revenue growth of 21% and adjusted free cash flow margin of 35% in its fiscal 2025 — has a strong track record of earnings beats since McDermott took over in November 2019. But ServiceNow's P/E ratio of 25 is saying forget about it, those estimates are unreasonable because ServiceNow will be destroyed by AI — even if its workplace automation software is really good. The jobs it does right now are thought to be too easily spoiled by Anthropic, so why buy shares of ServiceNow? CRM .SPX mountain 2016-02-18 Salesforce's stock versus the S & P 500 since February 2016. Finally, my "Squawk on the Street" co-host Carl Quintanilla said something to me on Wednesday morning that truly struck a chord. He told me that Salesforce's stock has underperformed the S & P 500 over the past decade. We have a small position in a stock that we should not own. You can't own a stock that long that underperforms. I have been a believer that the company's Agentforce platform could restart the growth here and I would be nuts to sell the stock. However, one sell-side analyst wrote Wednesday that they see "underwhelming" usage of Agentforce. This stock now sells at 14 times forward earnings. We have all become convinced that because customers pay by the seat and there are fewer seats because of AI, its non-Agentforce business is going to slow. I have no idea if that is true and was under an impression that Agentforce has amazing growth. If the analyst is right about usage, then this stock will go down from here. Salesforce is a nightmare. It has so much promise, and I think Agentforce is a fantastic product. But it just doesn't matter. The village, the community, the market says the Anthropics of the world can enable companies to easily develop their own customer relationship management software. It's a small position. I don't want to lie awake at night thinking about it. But I do. Why not sell it? Because I still believe in Marc Benioff. I still believe in their model. In my own business career, I have used Salesforce twice to astounding results. I just can't leave it. I just hope that those aren't my famous last words. (Jim Cramer's Charitable Trust is long DHR, COF, GS, PANW, CRWD, MSFT, AMZN, META, GLW, NVDA and CRM. See here for a full list of the stocks.) As a subscriber to the CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. Jim waits 45 minutes after sending a trade alert before buying or selling a stock in his charitable trust's portfolio. If Jim has talked about a stock on CNBC TV, he waits 72 hours after issuing the trade alert before executing the trade. THE ABOVE INVESTING CLUB INFORMATION IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY , TOGETHER WITH OUR DISCLAIMER . NO FIDUCIARY OBLIGATION OR DUTY EXISTS, OR IS CREATED, BY VIRTUE OF YOUR RECEIPT OF ANY INFORMATION PROVIDED IN CONNECTION WITH THE INVESTING CLUB. NO SPECIFIC OUTCOME OR PROFIT IS GUARANTEED.
