Constellation Energy and Google, Marvell's AI Forecast, Option Care, Constellation Brands

Good evening. This is Trade and Ticker. Tuesday, October sixth, twenty twenty-six. Your evening company deep dive after the close.

Four names tonight, and two of them happen to share a name without sharing anything else. First, Constellation Energy, the country's biggest nuclear power operator, which jumped about twelve percent after signing a huge long-term power deal with Google. Second, Marvell Technology, the chip designer that just told Wall Street it sees a path to as much as ninety billion dollars a year in revenue by the end of the decade. Third, a smaller health care story with a big buyer behind it: Option Care Health, the home infusion company that McKesson and a private equity firm agreed to take private, sending its shares up about a third. And fourth, Constellation Brands, the beer company behind Corona and Modelo, which reported results right after the closing bell. No relation to Constellation Energy, by the way. One sells electricity, the other sells beer. One power deal, one A-I chip forecast, one take-private, and one beer report. Let's get into it.

A quick look at the board first. It was a broad rally, and the S-and-P five hundred finally got its record. The S-and-P finished about six-tenths of a percent higher at an all-time closing high, its first record close since August and its twenty-eighth of the year. The Nasdaq finished about half a percent higher, also at a record close, its second in a row. The Dow finished about half a percent higher, up roughly two hundred fifty points, but it is still a little more than five percent below its own record from early August. Ten of the eleven S-and-P sectors rose. Utilities led the way, and health care was the only sector that finished lower. Smaller companies lagged the big ones. So the shape of the day was simple: the large names did the lifting, and power producers were out front.

What made room for the rally? Bonds and oil finally gave stocks a break. On Monday, the ten-year Treasury yield hit its highest level since April of two thousand two. Today it eased, falling about four basis points to around five-point-two-seven percent. Oil steadied after weeks of war-driven swings. U.S. crude settled essentially flat at about eighty-nine dollars and forty-four cents a barrel, and Brent settled a little higher at about one hundred dollars and fifty-eight cents, after the Group of Seven countries agreed to release emergency stockpiles of crude and diesel. When oil calms down, inflation worries ease, yields come in, and stocks get room to run. That has been the pattern for a couple of weeks now.

On the Fed, traders now see only about a one-in-five chance of another rate hike at this month's meeting, down from roughly even odds a week ago, based on the CME FedWatch tool. On the data front, the U.S. trade deficit widened to one hundred five-point-six billion dollars in August, as imports hit a record high. Part of that is companies importing equipment for the A-I buildout, which tells you how much money is still flowing into data centers. And with third-quarter earnings season starting next Tuesday with the big banks, analysts expect S-and-P five hundred profits to jump about thirty percent from a year ago, with technology earnings seen up about two-thirds. That is the setup. Let's start with the biggest mover in the S-and-P.

Constellation Energy and Google.

If you have ever wondered where all the electricity for A-I is going to come from, today's deal is one answer. Constellation Energy, based in Baltimore, is the largest nuclear power company in the United States and the largest private-sector power producer in the world. It runs a fleet of reactors across the Midwest and the Mid-Atlantic. And on Tuesday morning, it announced a sweeping long-term agreement with Google.

Here is how the deal works, because it has two parts, and both matter.

The first part is new power. Google signed a twenty-year agreement to buy eight hundred ninety megawatts of electricity that does not exist yet. Constellation will create it by upgrading eleven of its existing nuclear units in Illinois, Pennsylvania, and New Jersey. The industry calls these upgrades uprates. Think of it like tuning up an engine you already own. Constellation will put in new turbines, steam generators, and digital controls so the same reactors squeeze out more power. Eight hundred ninety megawatts is roughly what one large new conventional reactor would produce. Constellation says the first uprate should be delivered by twenty twenty-eight, and it expects to invest more than four-point-three billion dollars to make it happen, creating about seventy-two hundred construction jobs along the way.

The second part is a commitment to the power that already exists. Google also signed a fifteen-year supply agreement for another twenty-seven hundred megawatts from Constellation's existing fleet in the same regional grid, known as P-J-M. That power is not tied to one specific plant. Its job is to give Constellation's operating plants long-term revenue certainty. Add the two pieces together and Google has contracted for about thirty-five hundred ninety megawatts. New nuclear capacity is only about a quarter of that. The rest is a long-dated promise to keep buying what Constellation already makes.

There is a third piece, too. Constellation is expanding a five-year technology deal with Google Cloud, using its Gemini Enterprise A-I tools to speed up the uprates and run its plants. So Google is both the customer for the power and a supplier of the software.

Why does this matter so much? Start with the grid itself. P-J-M is the largest power grid in the country, serving about sixty-seven million people from the Mid-Atlantic into the Midwest. It has been under real strain as data centers multiply, and there has been a growing fight over who pays for new power plants and transmission lines. This deal is built to answer that. Google is funding new supply instead of just pulling power away from everyone else. The companies also say the deal includes demand flexibility, meaning Google can curtail non-critical power use when the grid is under stress, like on the hottest summer afternoons.

Then think about speed. Building a brand new nuclear plant takes many years, and connecting any new plant to the grid can take years more. Uprates skip most of that. The reactor is already there, already licensed, already connected. That is why uprates and restarts of closed plants have become the fastest way to add firm, around-the-clock, carbon-free power. Google says that with this deal, it has now helped bring more than one-and-a-half gigawatts of new nuclear capacity onto the U.S. system from uprates and restarts.

Constellation's chief executive, Joe Dominguez, called the agreement a model for how technology companies and the energy industry can work together, with grid-wide benefits funded by private companies. And it fits a pattern. Constellation signed a deal with Microsoft in twenty twenty-four tied to restarting a reactor at the Three Mile Island site in Pennsylvania, and it reached a nuclear power deal with Meta in twenty twenty-five. Google now joins that list in a big way.

Now the stock. Constellation Energy jumped about twelve-point-three percent on Tuesday, at the top of the S-and-P five hundred leaderboard, and closed right around three hundred dollars a share. At one point it was up close to fifteen percent. And here is the context that makes that move interesting. Even after today, the stock is still lower for the year. Nuclear stocks were one of the hottest trades of twenty twenty-five, and that momentum faded this year. A twenty-year contract from one of the richest companies on earth is about as concrete as it gets.

The rally spread. Other power producers with big nuclear fleets, including Vistra and Talen Energy, rallied on the read-through, even though they are not part of this deal. Traders are betting that if Google will pay for this kind of long-dated, firm power from Constellation, other tech giants will want the same thing from the rest of the industry. Utilities were the best sector in the market today, and that was a big reason why.

Into the coming months, watch three things. First, execution. Uprates are faster than new reactors, but they are still big engineering projects across eleven units, and the first one is not due until twenty twenty-eight. Second, whether Vistra, Talen, or other nuclear owners land similar long-term deals with hyperscalers, because today's rally in those names is built on that hope. Third, the politics of power prices in P-J-M. Deals that bring new supply instead of just absorbing existing power are exactly what regulators have been asking for, and this one could become a template.

Constellation Energy in one breath: about thirty-five hundred ninety megawatts contracted by Google, roughly a quarter of it brand new nuclear power from upgrading reactors it already runs, more than four billion dollars of new investment, and a stock that jumped about twelve percent to lead the S-and-P. That is the first deep dive.

From the power plants that feed data centers to the chips that move data around inside them. Marvell.

Marvell Technology.

Marvell is not a household name, but if you have used anything that runs in the cloud, you have probably used its chips. Marvell designs the parts that move data, inside servers, between racks, and across entire data centers. It also builds custom chips for the biggest cloud companies, which increasingly want their own A-I processors instead of relying entirely on Nvidia. Its chairman and chief executive is Matt Murphy, and on Tuesday the company held its investor day in New York.

Investor days are where companies tell you where they think they are going over the next several years. Marvell went big.

The near-term headline: Marvell raised its forecast for fiscal twenty twenty-eight revenue to about twenty billion dollars, up from about eighteen billion dollars as recently as August. Analysts had been expecting about eighteen-point-two billion. The company said the extra two billion dollars comes from demand for its data center connectivity products, things like optical links and networking switches that connect A-I chips together. It still expects revenue of about twelve billion dollars this fiscal year. So going from about twelve billion to about twenty billion would be growth of roughly two-thirds in a single year.

Then the long-term headline, and this is the one that got attention. Marvell laid out a path to between seventy and ninety billion dollars in annual revenue by fiscal twenty thirty-one. To put that in perspective, the midpoint of that range is eighty billion dollars. A small group of analysts polled by Visible Alpha had been expecting fiscal twenty thirty-one revenue of about forty-seven billion. So Marvell is pointing to a business close to seventy percent bigger than what Wall Street had penciled in, roughly five years out.

Where would all that come from? Two places, mostly. The first is what Marvell calls interconnect. As A-I systems get bigger, the amount of data moving between chips explodes, and the old copper wiring can only carry it so far. That pushes the industry toward optical connections, using light instead of electricity, and Marvell makes many of the chips that make those optical links work. The second is custom silicon, the made-to-order chips Marvell designs for cloud giants like Amazon and Google. Marvell also raised its target for custom chip revenue a few years out.

Google is a big part of that custom story. In August, Marvell disclosed an agreement with Google that could generate up to one hundred twenty billion dollars in sales through fiscal twenty thirty-three, if performance milestones are met. That is not a guarantee, but it shows how much these cloud companies are willing to commit to having their own chips.

How did the stock react? Marvell rose as much as about ten percent during the day and finished about five-point-eight percent higher. That might sound modest next to a forecast like that, but remember the starting point. Marvell's stock has more than tripled this year. A lot of good news was already in the price. Rival Broadcom, which also builds custom A-I chips for cloud companies, rose in early trading too.

And Marvell was not the only chip story pointing the same direction today. A-M-D rose about two-point-eight percent after its chief executive, Lisa Su, said the company plans to substantially increase its chip supply in twenty twenty-seven to keep up with A-I demand. She said it plainly to reporters in Taipei: "We've been able to increase our supply as we've gone through twenty twenty-six, and we're going to substantially increase our supply in twenty twenty-seven." She said A-M-D needs more advanced wafer capacity, is working closely with memory makers to secure supply, and is now planning its supply chain three to five years ahead. She was in Taiwan meeting manufacturing partners including Foxconn and was due to meet Taiwan Semiconductor. A-M-D's market value recently topped one trillion dollars.

Put Marvell and A-M-D side by side and you hear the same message from two different corners of the chip business. Demand is not the problem. Supply is. The companies are not talking about whether A-I spending lasts another year. They are reserving factory capacity and memory for years ahead.

There is a flip side. When a stock has tripled and the company raises its long-term targets this much, the bar for every future quarter goes up with it. Forecasts five years out are goals, not contracts, and much of this growth depends on a handful of very large customers.

Into the coming months, watch three things. First, whether Marvell's next few quarterly reports actually track toward that twenty billion dollar target for fiscal twenty twenty-eight. Second, how much of the custom chip story turns into firm orders, especially from Google. Third, the supply chain. If A-M-D, Marvell, and Nvidia all need more advanced wafers and more memory at the same time, the bottleneck shifts to the factories, and that is where costs and delays show up.

Marvell in one breath: fiscal twenty twenty-eight revenue now seen around twenty billion dollars, a path to seventy to ninety billion dollars by fiscal twenty thirty-one, a stock that finished almost six percent higher after a huge run this year, and A-M-D saying the same thing about demand from the supply side. That is the second deep dive.

Now from the most crowded trade in the market to a quieter corner of health care, on a day when health care was the only sector that finished lower. Option Care Health.

Option Care Health and McKesson.

Option Care Health is the smaller company tonight, but the story behind it is a big one. Option Care is the largest independent provider of home and alternate-site infusion services in the country. Infusion means medicine delivered through an I-V, often for complex or chronic conditions. For years, most of that happened in hospitals. Option Care's whole business is moving it to patients' homes, or to smaller infusion centers in the community, where it is usually cheaper and more comfortable. The company works in all fifty states, has about eight thousand employees, including more than five thousand clinicians, and serves more than three hundred thousand patients a year.

On Tuesday morning, McKesson, the giant drug distributor, and the private equity firm Clayton Dubilier and Rice agreed to buy Option Care for thirty-two dollars and five cents a share in cash. The deal values the company at about five-point-eight billion dollars including debt. That price is about a thirty-seven percent premium to Monday's close. The Financial Times had reported on Monday that a deal was close, and the stock had already started moving after hours. On Tuesday, Option Care shares soared about thirty-three percent and finished around thirty-one dollars, roughly a dollar below the offer price.

The structure is the interesting part. This is not McKesson buying the whole company outright, at least not yet. Clayton Dubilier and Rice will own about fifty-one percent. McKesson will invest about one-point-four billion dollars for the remaining forty-nine percent. And the agreement sets up a framework for McKesson to buy out the private equity firm's stake at some point in the future, subject to conditions and regulatory approvals. Option Care will stay a separate company with its own management team, and once the deal closes, its stock will no longer trade on the Nasdaq.

Why structure it that way? Think of it as McKesson buying an option on a bigger move. It gets a big seat at the table now, shares the cost and the risk with a private equity partner, and keeps a path to full ownership later. For McKesson, it also means it will record its share of Option Care's profits through what accountants call the equity method, rather than folding all of Option Care's revenue into its own results right away.

The strategy is easy to follow. McKesson's chief executive, Brian Tyler, said the company wants to invest where its skills can improve access to complex therapies in lower-cost settings, at or closer to home. McKesson already has a fast-growing oncology and multispecialty business. That segment, which includes infusion services, brought in about fourteen-point-two billion dollars of revenue in its latest quarter, up about a third from a year earlier. McKesson also already runs a large infusion network in Canada. Option Care would give it a national footprint in home infusion in the United States.

Analysts liked the fit. A J-P Morgan analyst, Lisa Gill, said McKesson's focus on specialty drugs and community care makes the deal appealing, because it can bring its supply chain strengths to Option Care's complex chronic therapies. A Leerink Partners analyst, Michael Cherny, said the deal lines McKesson up with what the industry calls the site-of-care shift, the long move away from expensive hospital settings.

And that shift is the big story here. The population is aging. More of the newest drugs are complex specialty treatments. Insurers and employers want the cheapest safe place to deliver them, and patients would generally rather be home. Private equity has been buying into home health for exactly that reason. This is another private equity buyout of a home health provider, after Enhabit.

Option Care is withdrawing its financial guidance and will skip the conference call with its November fourth results. The deal needs a vote from Option Care's shareholders plus regulatory approval, and the companies expect it to close in the first half of twenty twenty-seven.

Into the coming months, watch three things. First, that gap between Option Care's share price and the thirty-two dollar and five cent offer. Right now it is roughly three percent, which tells you the market sees a good chance this closes, with some room for regulatory risk and the long timeline. Second, the antitrust review, since a major drug distributor taking a big stake in the largest independent infusion provider is the kind of thing regulators look at closely. Third, whether this sparks more deals across home-based care, as other distributors and health systems try to keep up.

Option Care in one breath: thirty-two dollars and five cents a share, about five-point-eight billion dollars including debt, a stock up about a third and sitting just below the offer, McKesson taking about forty-nine percent with a path to the rest, and a close expected in the first half of twenty twenty-seven. That is the third deep dive.

And finally, the report that just came out after the closing bell. Constellation Brands.

Constellation Brands.

Constellation Brands is the company that sells Corona, Modelo, Pacifico, and Victoria in the United States. It also owns wine brands like Kim Crawford and Robert Mondavi, and spirits like High West whiskey. But beer is the whole ballgame here. Beer is the vast majority of its sales and profits, and for the past year or so, investors have been worried about it.

Constellation says its brands have the strongest following among Hispanic consumers of any major U.S. beer supplier. But beer demand has been under pressure, and so has the stock. So tonight's report was not really about one earnings number. It was about whether beer is stabilizing.

Let's start with the headline numbers for the quarter that ended in August. Net sales rose about six percent to two-point-six-three billion dollars. Comparable earnings per share, which strips out one-time items, came in at three dollars and seventy-four cents, up about three percent from a year ago. Reported earnings per share were three dollars and thirty-two cents.

Now the part that matters most, and it takes a minute to explain. Beer sales rose about five percent to about two-point-four-seven billion dollars, and shipments rose about five-and-a-half percent. That sounds healthy. But there is a second number the industry watches even more closely, called depletions. Shipments are what Constellation sends to its distributors. Depletions are what those distributors actually sell on to stores, bars, and restaurants. Depletions are the closest thing to real consumer demand. And depletions fell about six-tenths of a percent.

So why did shipments rise while depletions fell? Distributors were restocking. Constellation says it spent much of the first half of the year catching up after distributors ran their inventories low, and that inventories are now in healthier shape, though still below normal levels. That means part of this quarter's sales growth came from filling the pipeline, not from more people drinking more beer.

Look inside the depletions and you see the real tension. Modelo Especial, the company's biggest brand, saw depletions fall about two percent. Corona Extra fell about five percent. Those declines were partly offset by strong growth in smaller brands. Pacifico was up about nineteen percent, Victoria about fifteen percent, and the Modelo Chelada line about five percent. The company also blamed weaker than expected sales in stores around the World Cup in June and July.

There is good news in here too. Constellation says it was the number one gainer of market share in the entire U.S. beer category during the quarter, picking up more than eight-tenths of a point of share, and that it had five of the top fifteen brands gaining share. Modelo Especial is still the number one beer brand in the country by dollar sales. In other words, Constellation is winning a bigger slice of a beer market that is not growing.

The cost of winning that share showed up in profits. Beer operating margin fell one-point-six percentage points to thirty-nine percent. Lower tariff costs helped, but that was more than offset by higher spending on marketing and overhead. Constellation is spending to defend its brands, and that eats into margins.

Wine and spirits, which used to be the problem child, actually had a good quarter. Net sales in that business rose about seventeen percent, and depletions rose about ten percent, helped by Kim Crawford and Mi Campo tequila.

What about the full year? Constellation kept its outlook for comparable earnings per share at eleven dollars and twenty cents to eleven dollars and ninety cents. It raised its reported earnings outlook to eleven dollars and eighty-five cents to twelve dollars and fifty-five cents. And it still expects beer net sales for the full year to land somewhere between down one percent and up one percent, with beer operating margin between thirty-seven and thirty-eight percent. That margin target is below the thirty-nine percent it just posted, which tells you management expects spending to stay high in the back half of the year.

The company also kept returning cash. It returned about four hundred million dollars to shareholders in the quarter through buybacks and dividends, and it has repurchased about five hundred thirty million dollars of stock so far this fiscal year.

So how should you read this? The money side held up: sales grew, earnings grew a little, and the full-year target held. The demand side is not fixed: depletions slipped, and Modelo and Corona are both shrinking at the store level. The bull case is that share gains and healthier distributor inventories set up a better second half. The bear case is that the company is paying more in marketing just to stand still.

Into tomorrow, watch three things. First, the conference call at eight o'clock Eastern with chief executive Nicholas Fink and finance chief Garth Hankinson, and whether they say anything about depletion trends in September and October. Second, any comment on the Hispanic consumer and on how much marketing spending will rise from here. Third, how the stock trades once the regular session opens and investors have had the full night and the call to weigh it.

Constellation Brands in one breath: sales up about six percent to two-point-six-three billion dollars, comparable earnings of three dollars and seventy-four cents, beer depletions down about six-tenths of a percent with Modelo and Corona both lower, market share still rising, and a full-year earnings outlook that held. That is the fourth deep dive.

So let's bring it all together. The S-and-P five hundred closed at a record for the first time since August, and lower yields and steadier oil gave stocks room to breathe. Underneath that, the A-I buildout is no longer just a chip story. It is a power story, with Google signing up decades of nuclear output from Constellation Energy, and a supply story, with Marvell and A-M-D both saying demand is running ahead of what they can build. Away from A-I, McKesson is betting on care moving into the home. And Constellation Brands showed what it looks like to gain share in a beer market that is not growing.

Here's what to watch into tomorrow morning. Constellation Brands holds its call at eight o'clock Eastern, and that will shape how the stock opens. At two o'clock Eastern, we get the minutes from the Fed's September meeting, the one where it raised rates, and investors will be looking for any hint about whether another hike is coming. And we will be watching whether the ten-year yield keeps easing after hitting a twenty-four-year high on Monday. We will have the futures board and all of that for you on tomorrow morning's show, before the opening bell.

That is your evening company deep dive. Enjoy the rest of your night. Thanks for listening. I'm Trade and Ticker.