We have so many struggles going on right now that it’s hard to believe the market can stay up at this level for much longer. It’s a visceral moment where the losses mount quickly if you are on the wrong side of the trade. We all know that it’s time to embrace the bear, right? But we always recall that when we declare we cannot take it anymore and sell — whether it be in the days after “Liberation Day” or the days leading up to last week’s Iran war truce — we’re immediately proven wrong. The result: The selling that can go from a trickle to a flood stays at a trickle longer than we think, aided, of course, by a brain-dead bond market that doesn’t seem to care about the fundamentals, even as that’s all it is supposed to care about. The benign bond market, though, masks some cogent themes. They need to be looked in to. So, let’s do this. Let’s examine the three most salient battlegrounds in play right now: Iran, software and earnings, not because they are emblematic of the market, but because there is nothing emblematic about this market at all. Instead it is just tugs-of-war that are surprisingly disparate and surprisingly separate from each other. Iran war First, we know the Iran war basically caused oil to double from where it started the year — that is, before last week’s big decline on the two-week ceasefire news . Nevertheless, if history was our guide , the S & P 500 should be down some 20% from its highs right now. Not only is history being disobeyed, but since the market rally began in all of it pessimistic glory two weeks ago, we are capable of punching to new highs with another strong week. The S & P 500 ended Friday 2.3% below its Jan. 27 all-time closing high. At its lows of the year on March 30, the index was off 9% from the highs. Why is that? First, we use less oil than we used to for a variety of industries, including the refining of gasoline. Interestingly, it is not our continental self-sufficiency that has helped keep down the price of oil or gasoline; both are set by the world’s market forces. In the past, oil shocks have primarily hurt us at the pump. But today’s vehicles are more fuel efficient. Perhaps more important, despite all of the griping, it simply isn’t that high versus inflation. We’ve been here before and it didn’t cause a downturn, so the market is concluding it won’t do so again. Additionally, natural gas is the largest single source of power generation in the U.S. We’re blessed that we produce a ton of it, and that its domestic price is less connected to global market forces than what we see with oil. Since the Iran war began on Feb. 28, U.S. natural gas futures are actually down over 7%. The chart of Europe’s benchmark natural gas contract looks much different . While the U.S. is exporting a lot of natural gas via LNG terminals, we still have plenty of domestic supply. It’s obvious that our country’s efforts to become more fuel efficient and energy independent have worked. So why does the price matter so much? I think there’s a tremendous fear that oil is about to go to $150 when the various strategic petroleum reserves around the globe run out — a big reason for the urgency of the unsuccessful peace talks in Pakistan this weekend. If that happens, we will most surely see some countries in real trouble, like South Korea, Japan, Taiwan and many others, especially in Europe, where the price at the pump will be prohibitive. Ultimately, we will see some inflationary ripple effects here that could lift bond yields and make it so even President Donald Trump’s dovish nominee to lead the Federal Reserve, Kevin Warsh, can’t cut rates . The tone of our market will surely change if that happens. Friday’s March consumer price index report was already elevated by tariffs in key categories like apparel. Now, the oil-influenced part of the economy will cost more, and while we can asterisk both the tariffs and the war, the Fed doesn’t want to risk its credibility by cutting into one of the more serious inflation waves our nation as had. It’s unfortunate that housing and its accoutrements have only gotten weaker — a testament to how many long-term mortgages were taken out when rates were historically low during the Covid-19 pandemic . We don’t have home-price appreciation any more, but the depreciation where it does exist has not inspired transactions. The “lock-in effect” is real. Without intrusion from the bond market into the stock market, there is no uniform way for the active managers to do anything but shift their money from sector to sector. They can’t leave the market all together because when they do, you get the kind of rally we’ve had since the March 30 bottom — one that interest rates gave permission for. The yield on the benchmark 10-year Treasury note topped out on March 27 at almost 4.5%, following a big surge during the first few weeks of the Iran war. The 10-year yield settled Friday at 4.32%. Bond prices move inversely to yields. We have not made enough of the low rates as a spur for stock buying. Maybe that’s because these rates almost seem impossible to believe. Every session they stay up is another day money gets put to work. It’s almost as if they can withstand everything, yet all we talk about is the Fed and its minions even as they aren’t the cause of the low rates. I have been negligent in bringing up the power of low rates because it’s the chief reason the bulls keep winning and why the market isn’t down more. Let’s not overthink that. And let’s stop trading on the discourse of Fed heads or the possibilities of rate cuts versus rate hikes. Sure, low rates may be the endpoint of a million different decisions, but it is the endpoint and we should not look a low-rate gift horse in the mouth even if we think it’s a Trojan horse. Tech trade The second battleground is what’s happening with technology stocks. Here, we have an unheard-of roiling that went existential last week : the extinguishing of software at the hands of hardware and artificial intelligence. Given that the whole story of technology stocks since 1986 is the wasting of hardware by software, what’s gone on this year is as astounding as it is frightening. We have software stocks that are down 30% to 40% despite the fact they’re making incredible amounts of money. We have hardware stocks up 50% to 150% with no sign of stopping. It’s breathtaking. The attack on software is so complicated that I will do my best to unpack it, but there’s no surety anyone can comprehend something this hideous. Let’s start with the software-as-a-service (SaaS) companies. Here, we have a once-great business model that is said to be broken because the companies that use SaaS vendors supposedly are no longer growing. They don’t need more SaaS because AI has allowed these companies to cut back on hiring, and AI coding tools allow you to create in-house applications that mimic what you previously paid for externally. Now, there is no evidence that any of this actually occurring and many of the companies with stocks being eviscerated are still growing at a healthy clip. However, it doesn’t seem to matter in the eyes of investors. If there’s a belief that AI can hurt you, then it gets extrapolated rather quickly. It all seems alleged, not factual, but it doesn’t matter. We just know to dump and bet against the stocks that are — or will be — hurt by the power of AI to change things. The two biggest and most important allegedly crushed companies are ServiceNow and Salesforce . Neither has copped to any pain from AI, just pleasure. They laugh at the destruction thesis, and they say they haven’t even seen a slowing. There couldn’t be more of a disparity between what the Street thinks of these companies and what the managers think of their own businesses. These were — and, to some degree, still are — large companies. ServiceNow has an $86 billion market cap, but it peaked at almost $242 billion in January 2025. Salesforce clocks in at $152 billion, down from its December 2024 peak of $352 billion. These are highly visible companies that Wall Street had loved for years, particularly ServiceNow. NOW CRM,.SPX 1Y mountain ServiceNow’s stock performance compared with Salesforce and the S & P 500 over the past 12 months. We have owned Salesforce seemingly forever, although it is our smallest position and we have not encouraged buying it into this AI disruption sell-off. I am a huge believer in CEO Marc Benioff and have used Salesforce’s applications twice at my old stopping grounds, TheStreet.com. While it is expensive and hard to implement — we had to hire an outside contractor to install it — the product worked. We had a 30% lift in sales each time we introduced the product. I have heard gripes about its costs, but I have never heard anyone complain about the results. The most recent addition to its “forces” — Agentforce — is seeing traction, with annual recurring revenue of $800 million as of its late February earnings report. Major companies with huge customer-facing businesses swear by it. The company’s Slack is arguably the preferred way to communicate with coworkers. And it doesn’t matter one whit. Nor did the debt-fueled $25 billion accelerated share repurchase executed in mid-March, part of a larger $50 billion buyback authorization. That hasn’t kept the stock afloat at all, with shares touching a fresh 52-week low Friday. The whole escapade is a little crazy. Salesforce is one of the most successful companies in history. Last fiscal year its revenue topped $40 billion, and it has a record of beat-and-raise quarters challenged by very few. And yet, whatever it does, the Street thinks not only that its growth will slow — heck, it’s trading at 12 times forward earnings, per FactSet data — but some smart minds think the company may not survive . That’s right, survive. I don’t believe that to be the case. But my belief has proven to be ineffective in the face of an avalanche of selling in the stock. It is becoming harder and harder to justify holding even this tiny piece of what I still think is a tremendous company. The gulf between the business and the stock is simply shocking to me. Again, it doesn’t matter. If I were to return to my hedge fund days where Karen Cramer ran the trading desk, I can imagine exactly what would happen. She would be furious that I had made her ride it down to these levels, turning a big gain into nothing. She would lose so much patience that she pulled the ultimate trick reserved for when she was really upset with my buffoonery. “Would you mine going downstairs and getting me a soft pretzel from the guy at Wall and Water?” she would say. “And don’t forget the mustard!” When I got back she would say, “Thanks. Oh, and while you were gone, I sold the Salesforce position.” I would look up, angered, and she would spit out, “Hey, if you are so pissed off at me, you can always buy it back.” I never did. Telling. Yeah, it would be gone at Cramer & Co. It would have been gone a long time ago. My astonishment in its decline is trumped only by its relentless speed. I want to wait for the several-day rally I would expect from this extremely profitable company, if I am going to sell it, but it can’t seem to put together even the most minor of a win streak. Put it in the to-be-determined class of stock. The ServiceNow decline is even more odd. Business is very strong. The company has embarked on a full-scale remake, with CEO Bill McDermott recently telling The Wall Street Journal that 50% of its new business revenue is coming from its non-seat based pricing model — remember, it’s the seat-based licensing model that investors fear is cooked if companies cut their headcounts thanks to AI. But that just leaves the Street very worried about the other 50%, even as it is extremely fast growing and profitable. UBS just downgraded the stock Friday. For all intents and purposes, this ServiceNow is every bit as good or better than the one that had less AI. But it doesn’t matter. This is another company that the clients all laud. Any amount of homework makes you love it more. Doesn’t matter. Go get me a soft pretzel and then tell me you want to buy it back. You won’t. You will be relieved I took action while you remembered the mustard. There’s so much that is calamitous about this software unraveling — first in SaaS and now in all enterprise software — that I keep thinking if we didn’t own any of these, it would just seem to be one dark comedy. Take Palantir , Alex Karp’s brilliant brainchild, a data-analytics company that clients swear by. I now know it as the third-largest position in the iShares Expanded Tech-Software Sector ETF , often just called the IGV in reference to its ticker symbol. The IGV is the go-to way to short software and hedge other tech positions. It gets hit every time some manager says he needs to protect himself from AI. The IGV is down 29% year to date. Palantir knew AI before there was AI. All that bragging Karp does about Palantir obliterating the “Rule of 40” metric cannot protect Palantir’s stock because of its towering weighting in a foolishly designed index. It could have a “Rule of 100” and it would mean nothing. As of Friday, Palantir’s weighting in the IGV is 8.29% while its market cap is $306 billion. Microsoft, with its $2.75 trillion market cap, has an 8.9%% weighting in the fund (more on Microsoft in a moment). My point: Palantir is a much smaller company than Microsoft, but it does not have a much smaller weighting. Or consider Club names CrowdStrike and Palo Alto Networks , both in the IGV and both designed to protect companies from predators including AI-enabled predators. Their stocks are being sold just as hard as if they caused cyberattacks. Palo Alto has a 5.7% weighting in the IGV; CrowdStrike checks in at 4.4%. Oh, and then there’s true AI roadkill: Adobe and Atlassian , both of which were almost meant to be destroyed by AI. If you ask me to design two companies that could be obviated by AI, it would be those two. According to FactSet, Photoshop maker Adobe trades at 9 times forward earnings and still has a $91 billion in market cap. Where is that market cap going to go? Atlassian, which makes collaboration software, is worth $15 billion. It feels like a software version of Digital Equipment, the late maker of minicomputers. These stocks can’t catch a break. None of them. Yes, the year-to-date declines are legit in Adobe (down 35.6%) and Atlassian (down 65%), but all of the others? Let’s just say their declines are way out of whack with the truth. They deserve better, yet it means nothing. The flipside, the hardware stock sector, is even more insane. After being despised for ages and ages, these stocks have become scarce gems that you buy and hold seemingly no matter what. Consider memory and storage, where AI has created a massive wave of demand for their products. The big four: Sandisk , Seagate , Western Digital and Micron . Only Micron is really taking advantage of this moment and building out a lot of manufacturing capacity. The rest are tight-floated monsters that just keep reporting better and better earnings because they keep raising prices. Of course, we’re a long-only portfolio here at the Club. But, if I were at my old hedge fund, I can tell you that we would own deep-in-the-money calls on all four because the earnings estimates are probably still too low and there are too many shorts in them. We wouldn’t buy the common stock because they are pure momentum names now that tend to get hit with nasty declines, just like we saw when Micron reported, or when we read that Google has some propriety algorithm that could create less memory need — a knee-jerk reaction that has since been declared misguided. Then there’s Intel . Left for dead after former CEO Pat Gelsinger wrecked the balance sheet, it is now buying back the 49% stake in an Irish chipmaking facility that it had desperately sold to raise quick cash to Apollo in 2024. That’s a clear sign of renewed strength. INTC 1Y mountain Intel’s stock performance over the past 12 months. What happened to Intel to ignite its stock over the past year with brilliant new CEO Lip-Bu Tan at the helm? Turns out, the kind of agentic AI computing enabled by Nvidia’s graphics processing units (GPUs) also requires lots of central processing units (CPUs) to help things run smoothly and efficiently. CPUs have long been Intel’s bread and butter, and they still make one of the best options available for data centers. Another driver of the stock: Intel is intensely involved in advanced semiconductor packaging, a very lucrative part of the data center food chain. That makes a ton of sense considering that, from his time at Cadence Design Systems , Tan knows semiconductor packaging better than anyone on Earth. Just as it seems there is no price you can’t sell Salesforce at and be unhappy, there’s no price you can’t pay for Intel. I am accepting that Intel should never have been at $20 a share, not that it shouldn’t be at $62 and change. Actually, it should be at $70. There are so many data center innards that know no bounds either. Director of Portfolio Analysis Jeff Marks and I sit in the office and marvel at, well, Marvell Technology , or AMD , which have soared 46% and 25%, respectively, since the March 30 market bottom. Same goes for anything fiber and laser like Lumentum and Coherent , both of which inked strategic partnerships with Nvidia this year. Ciena is the optical backbone. Thank heavens for Club names Corning and Qnity . Corning is our play on fervent demand for fiber optics inside data centers. Spun off from Dupont last year, Qnity supplies all sorts of materials used to fabricate and package semiconductors. The bounty extends to our positions in GE Vernova and Eaton . It continues to Caterpillar and Vertiv. While these four companies may not be technically considered tech stocks, they are integral parts of the broader AI trade because their products are indispensable to the infrastructure buidout. It goes to Applied Materials , Lam Research and KLA Corp. And, of course, there is CoreWeave , the pure-play AI computing provider. I don’t think I have ever seen such a ferocious sector move in my life, made even more vicious by how few stocks there really are in the group versus the software stocks. There is a hardware stock shortage for certain. Oddly until last week, the stocks of the guys who pay the bills, like Amazon and Meta , had performed incredibly poorly. Same for Nvidia, which couldn’t go anywhere. It’s disconcerting when the most obvious go up last. But I console myself by saying that playing catch-up is not a one-week affair. Before we move on from the software-versus-hardware dichotomy and journey to earnings, let’s consider the biggest conundrum of all: What do we do with the stock I affectionately used to call Mr. Softee? No stock is more of a quandary than Microsoft. MSFT 1Y mountain Microsoft’s stock performance over the past 12 months. Microsoft has become this market’s pitiful helpless giant. It can’t seem to do anything right, but because it is Microsoft you can’t even say that aloud. It’s obvious — perhaps to all but management — that Copilot isn’t in the same league as the tools from OpenAI or Anthropic. At one time, we would have thought that Microsoft was OpenAI. Now there’s all sorts of questions about their relationship, though, it should be noted, OpenAI CFO Sarah Friar told me last month that Microsoft is “an incredible partner.” For almost four decades, Microsoft was a must-own stock, perhaps the most “must-own” stock in history (to be sure, it did spend time in the wilderness after the dot-com bubble burst, like many other tech stocks of the era). That status now seems over. But before you blow it out of the position because it is software, remember it has so much cash it can save itself. Didn’t Google save itself? Aren’t they every bit as smart as the people at Google? Jeez, if Microsoft would simply buy a tremendous AI company, its stock could soar more than it paid for the target. There, sub rosa, I just gave you the debate that swirls through my head every time I look at our positions. Isn’t the report of Microsoft’s illness — not its obituary — premature? I think it is, but I worry so much that management doesn’t know it. I still think the odds favor betting on the company righting the ship. But that has to happen fast. The stock is gathering naysayers by the hour. I don’t think the software-into-hardware trend is going to change any time soon, even as we thought it would get a short-covering rally some time last week. Nope. Earnings season arrives Finally, earnings. It’s a bank bonanza this week . I think we own the best ones out there in Goldman Sachs and Wells Fargo — the only exception is Citi , but you can’t own them all. Goldman Sachs is the first big bank to report this time, which I think is terrific. The order shouldn’t matter, but in many past earnings seasons, Goldman suffers by comparison and doesn’t stand out because it is last to report among its peers. That’s over. I think Goldman has a better story to tell than any traditional bank, and this will be the quarter to tell it. You could fault us for not switching off the Wells Fargo horse and going to Citi, but I am skeptical of the Citi rally. The main reason this stock keeps going up is because people keep underestimating how horrible this bank was before Jane Fraser took over in March 2021. So, it beats the low-ball estimates every time. If the analysts were simply to post real estimates based on the new bank that Fraser has assembled, it wouldn’t go up much anymore. I keep fearing this will be the quarter that there will be more realistic estimates and the company won’t trounce them. That’s what keeps me out of it. The most exciting thing for me will be to hear the commentary about whether big banks are going to start buying little ones. At one time, no bank was supposed to own more than 10% of the nation’s deposits. But Bank of America and JPMorgan Chase exceed that. I think this group of antitrust regulators will allow our nation to begin to divvy up the regionals, allowing for growth and rationalization in an industry in bad need of it. I just need to hear it from the banks themselves. The only other stock that I really care about this week is Johnson & Johnson , and I think we will get our typical good quarter from our newest portfolio stock. Remember, though, the stock is a rocky trader four days a year (the days it reports quarterly numbers), while it is usually smooth sailing on most others. I have seen this stock be up $4 a share in premarket trading, then open flat before falling down $4, only to finish the day up $3. Be ready. Bottom line No matter what, remember, we are up at these exalted levels — and they are exalted — not because of earnings, not because of inflation, but because of interest rates, which have become ridiculously tame in recent weeks. If they stay benign, then earnings season will start off just fine and we’ll be right back into the thrill of hardware victory and the pain of software defeat by the end of the week. (See here for a full list of the stocks in Jim Cramer’s Charitable Trust.) 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.
Why the stock market has stayed resilient despite the Iran war and software stock rout
Apr 12, 2026
