6 Best Stocks to Buy Now in September 2026

September has historically been the worst month of the year for the S&P 500. Since 1948, the index has lost an average of around.7% during September, making it the only month with a meaningfully negative historical return. But this September presents investors with a much harder decision than simply buying or selling the market. When we take a look at stocks, they remain close to record highs. Some of the world's largest companies, they're priced for years of future growth and the bond market. That's presenting a warning that I don't think investors should ignore. If we take a look, long-term interest rates, they're close to their highest levels in nearly two decades, and it raises the required return for every stock we buy and makes paying the wrong price increasingly dangerous. However, sitting entirely in cash could prove just as costly. Corporate earnings remain strong. artificial intelligent investment continues to accelerate and several exceptional companies. They're already trading well below their recent highs. So, what I've done in today's episode, I've searched across the market for six companies I genuinely consider buying in September. One's fallen more than 20% this year. One's growing a critical AI business near 80%. Another trades 18 times forward earnings. And my number one stock has almost no downside under my cautious valuation, but 44% upside in my base case. Let's begin with the market risk every investor needs to understand. And September's weak record, it isn't based on one bad year across several decades is consistently produced weaker average returns than any other month. Doesn't mean the market must fall. Seasonality is a risk factor. Is not a prediction. and the danger that investors enter the month complacent, fully invested and completely unprepared for volatility. After summer trading volume decline, institutional investors return, companies issue more debt, portfolios are rebalanced, and the markets begin focusing on the final months of the year. And historically, the pressures often increase during the second half of September. That is useful because it means we don't need to deploy all our cash on the first trading day. A stock can be attractive today and still become more attractive 2 weeks later. If we look at the VIX where it's currently close to 15, that suggests investors expect relatively calm markets. Low volatility, therefore, it's not automatically bearish, but it means protection is inexpensive and expectations for disruption are low. The combination here makes any unexpected inflation, employment, or interest rate surprise more important. And the bond market is where I see the clearest near-term risk. The 2-year Treasury yield, that's around 4.2%, the 10 year around 4.7, the 30 years climbed above 5%. Even if the Federal Reserve eventually reduces short-term rates, it doesn't control the long end of the bond market. Long-term yields reflect inflation expectations, government borrowing, economic growth, and the amount of new debt investors are being asked to absorb. When investors can obtain around 5% from longdated government bonds, a stock trading at 30 or 40 times earnings must offer exceptional growth to justify the additional risk is why I'm not building today's list around assumptions that interest rates immediately collapse. For each stock, I want a credible return even if yields remain elevated. And higher bond yields also affect companies directly. They increase refinancing costs, reduce the present value of distant cash flows, and create competition for dividend paying stocks. That will be particularly important when we reach the two electricity companies in today's countdown. And the stocks that ultimately survive this environment, they need strong balance sheets, dependable earnings, or enough growth to overwhelm the higher discount rate. And we can see the broader market is also expensive. The shiller cape ratio is now around 42 times earnings placing it close to the 98th percentile of the historical range. Only the peak of the dotcom bubble was clearly higher. Now obviously it doesn't tell us when the market will fall. Valuation is notoriously poor at predicting what happens next week or next month. It does tell us though that future returns becomes increasingly dependent on earnings delivering what investors already expect. and retail investors. They've also returned aggressively to call options in several major tech companies. Again, doesn't automatically mean a crash is coming, but it does show that optimism is no longer scarce. More importantly, though, performance inside the markets been extremely uneven. Some technology leaders have risen while Meta and Tesla have suffered doubledigit declines. The dispersion here creates a much better environment for selecting individual companies than blindly buying everything at the same valuation. So the objective is not to abandon excellent businesses. It's to identify where fear, temporary uncertainty or slowing expectations have created a reasonable entry price. And despite the risk, this has been an exceptionally strong earning season. Technology companies alone added around 1.4 4 trillion in market value as the results came through. Which it does matter because sustainable bare markets are normally associated with deteriorating earnings, tightening credit or an approaching recession, not simply an expensive valuation. Where investors who wait for every risk to disappear, they usually end up buying after prices have already recovered. My approach is therefore not to predict the September crash is to preserve enough cash to benefit if volatility arrives while gradually buying businesses whose valuations already offer a degree of protection. And professional investors are making a similar distinction. They're not necessarily abandoning the market, but they are becoming more selective about where capital is deployed. And that leads directly into the own external clip that I'm using in today's episode. And listen carefully to how Jeff Kilberg describes the current market. His central point is not that September will be easy. It's that the widening gap between individual stocks is creating opportunities for those investors who are willing to be selective. >> Deploy cash to be a stock picker. This has been a phenomenal year. Joe, you think about look at the DSPX. That's a dispersion opportunity. If you look at the DSPX in July, it was 47, a really high rating. Typically, for the last couple years, it's been around 22. Today, it's coming in at 34 and a half. And all that means in layman's terms is that you are seeing a lot of dispersion between the stocks in the S&P 500. So that provides opportunity and I think if you look at software look at IGV compared to uh socks which was obviously the semiconductor ETF that dispersion was about 100%. That's narrowed in. So I think the time going in the fall seasonally September people get a little bit apprehensive but I think now is the opportunity and the risk is actually Joe being underinvested. >> That's the central argument behind today's countdown. I'm not buying the entire market indiscriminately, and I'm not betting everything on a crash. I'm ranking six individual companies using four factors: business quality, earnings growth, valuation, and the downside if my assumptions prove wrong. Let's begin with number six. And in sixth place is Next Era Energy, ticker NE. It's the least explosive company in today's list, but it offers a more defensive route into one of the decade's largest investment themes, America's rapidly increasing electricity demand. Now, the shares trading just below 82 odds, pretty much flat year to date. Over the last year, they're up around 14% and they currently remain around 17% below their 52- week high, sitting in fact around the mid to lower end of the 52- week range with a double of a weak buy from boasting Alpha as well as Wall Street. And the United States already has more data centers than every other major country shown here combined. Artificial intelligence is accelerating that demand. The Next Era is not a pure AI stock. It owns Florida Power and Light, one of America's largest regulated utilities alongside a major renewable energy development business. Now, this combination gives Next Era exposure to population growth, grid investment, renewable deployment, and rising demand from data centers. revenues growing around 11% today and analysts expect close to 12% forward growth and forward EBIT DAR as well as forward EBIT both of these are expected to approach around 20% although longerterm EPS as we can see here expected to grow more conservatively around 8% that is still attractive for a regulated utility but it doesn't justify paying just any price and NE well they report excellent margins a 51% Ebit margin 32% on the net income margin and a return of equity that sits above 17%. Now the most important figure here is in fact their lever free cash flow margin which is negative 62%. And next era is spending more capital expenditure than it currently generates in revenue. The investment though should create decades of future electricity production but it also means the company remains dependent on debt markets and external financing. higher interest rates therefore mass enormously and at around 19.7 times Ford earnings. Next era trades below its 5year average of 23. Its 3% yield is also above the 5year average of 2.7. It makes the shares more attractive than they were several years ago, but I wouldn't call it obviously cheap. And then if we get to the blue tunnel from Simply Safe Dividends, which highlights fair value intrinsic price, we can see pretty much for the first time since September, so nearly a year ago, this one sitting just below the bottom end indicating a potential undervaluation signal. Look at the last 5 years, we actually notice this one. If you go all the way back to the end of 2023, it was in a severely undervalued level. Now, whilst not as much of a disconnect, it's still highlighting potential undervaluation. And we can see in our valuation model specifically for the multiples valuation we get a value of $76 implying around 7% downside. The dividend model comes to around $17 implying more than 40% and averaging these methods we get to $96. But obviously these figures aren't exactly precise and in combination equates to a 16% margin of safety. Wall Street's average target comes to $98, representing 20% implied upside, but the range well extends from 55 on the lower end to 114. It demonstrates how dependent Next Era's value is on financing costs and long-term growth. My conclusion is that Next Era is a cautious income oriented buy, not a stock that I would chase. I become much more interested below 80 and considerably more interested between $72 to $78. So next era earns sick place because the valuation has improved, the dividend is attractive and electricity demand should rise. But high capital expenditure and interest rate sensitivity prevents it from ranking higher. Number five though provides much more growth but also comes with a different cash flow problem. And just before I move on, I want to let you know that I released one weekly copy where we cover severely undervalued stocks as well as what's gone in the market over the last few days. You can click below, sign up, read all of these straight away. And at number five, we have Constellation Energy. The shares are down around 22% year to date. Over the last year, they're down around 13% and they sit around 1/3 from their 52- week high. Currently trading towards the lower end of the 52- week range. We can see a near strong buy from Wall Street, 4.45 out of five. C Alpha, give it a buy, four out of five. and the decline with the company, it makes the stock interesting because the fundamental demand story, well, that's continued strengthening. We have global electricity demand from data centers that's expected to quadruple over the next decade. AI chips receive most of the attention, but those chips cannot operate without reliable round-the-clock electricity and nuclear power is one of the few scalable sources which are capable of supplying it without intermittent generation. And if we look at their latest quarter, well, their revenue increased 23% to 7.5 billion. Power and power related revenue that contributed 5 billion while other revenue increase sharply. Although worth highlighting, net income did fall 39%. The divergence tells us revenue growth alone for CG doesn't capture the entire picture. And when we look at their growth, well, revenue that was up 26% year-over-year. Forward revenue expected around 16%. EIT are expected to grow almost 20 while long-term earnings per share coming in at the 16% region. These are outstanding figures for an electricity producer. But profitability that is more complicated. Margins have improved significantly from their 5year averages. Cash from operations positive and return on equity we can see just above 15%. But levered free cash flow margin negative 21%. It prevents us from using the same conventional DCF approach which we apply to assetike companies later in the list. And at 23 times forward earnings, Constellation trades close to its 5-year average. Therefore, despite the large share price decline, the stock isn't a traditional deep value opportunity. Investors still paying a premium for the expected electricity shortage. Where we do get a reasonable signal, although it is sitting towards the lower end of the blue tunnel, look at the last 5 years. In fact, for the majority, investors were more than happy to pay a premium. Today, as we said, right towards the bottom end. Now, the multiples valuation suggests around $242, around 12% below the current price. The dividend model, well, that gets to $370. And averaging these extremely different outcomes, gives us $36. But the average creates more confidence than the underlining evidence deserves. That also equates to a 10% margin of safety. Where analysts are considerably more optimistic, their average target is $348, representing around 26% upside. and even the lowest display target here of 290 that's slightly above today's price. Now the bull case straightforward electricity becomes the limiting factor for AI deployment. Nuclear assets become increasingly valuable and constellation signs long-term contracts at attractive prices. The risk is that investors have already capitalized years of the future demand into the current multiple today at around $275. I consider only a small starter position. My stronger accumulation range would be around the $240 mark. Constellation ranks above next era because its growth is stronger and it offers more direct exposure to the power shortage. It remains number five because the free cash for evidence and valuation are not yet strong enough for a large commitment. Our next company has none of those capital intensity problems. And at number four, we have MSCI. Unlike the previous two companies, MSCI does not build power plants, own transmission infrastructure, or require enormous annual capital expenditure. It sells financial data, indices, analytics, and investment tools that become deeply embedded into customer workflows. They are flat year to date. Over the last year, they're also flat. currently trading around the mid to low end of the 52- week range where we get a strong buy from Wall Street, weak buy from Seek Alpha, and just in their latest quarter, revenue was up 12%, recurring subscriptions that contributed 613 million, and net income reached 342 million, producing a remarkable 39% margin. and their economics. Well, their exceptional gross margin sitting at 83%, EBITDAR margin sitting at 59%, net income margin 41, and return on total capital that exceeds 30%. Very few publicly traded companies combine recurring revenue, pricing power, and margins of this quality. And revenue is expected to grow around 10% with forward EBITR coming in near 12. EBIT sitting around 13% and earnings per share close to 14. Now, it's not explosive growth, but it is highly valuable when combined with recurring revenue and limited capital requirements. We can also see their free cash flow per share that's increased from around.7 cents in 2005 to more than $20 in 2025, an annualized growth rate above 18%. MCI, on top of that, they've also reduced their share count from 112 million to 73. That 35% reduction means each remaining shareholder owns a progressively larger portion of the business. The shares now trade around 27 times Ford earnings compared with a 5-year average above 36. The yield of 1.44% is also well above its historical average. The stock's not collapsed but his valuation has compressed while the underlining business continued growing. And we can also see the undervaluation signal for MCI although it has been very constant over the last year. Look at the last five last 10 years. This one, in fact, up until 2022, the beginning of it was trading at a massive premium. Now, we've seen over the last near two years, this one's trading at a discount, whilst the underlying fundamentals only continue to rise. And my base ECF assumes 12% cash flow growth, an 8% discount rate, and 3% perpetual growth. Now, this produces a value of $684, which equates to around 20% upside. The lower 10% growth produces $584 which is pretty much around today's price while the 14% which is actually in line with their 5year keer lower than their 16. So you could argue maybe conservative the rate we've used 14 though that gives 798 indicating 40% upside. So this tells us the present valuation is reasonable rather than extremely cheap. The reverse DCF also something to touch upon here 9.7% that is what is currently needed to justify today's price. That is achievable, but MSEI must continue delivering. And using that middle rate, we get a 17% margin of safety. Wall Street's average, well, $690, closely matching my base case valuation, and implies around 21% upside. The primary risk is market sensitivity, volatility, and assetbased fees, and around 6.5 billion of debt. But because most revenues recurring, and MSI's products are deeply embedded, this remains one of the highest quality businesses in the entire list today. I'd consider gradually buying MCI around $57 with greater conviction below $540. It ranks fourth only because the next three companies offer either a wider margin of safety or substantially greater upside. And number three is Broadcom and this is where the potential returns and the risks both increase sharply because Broadcom's become one of the most important infrastructure suppliers behind AI spanning custom accelerators, networking, connectivity, and enterprise software. They're up around 7% year to date over the last year up near 20% trading towards the lower end of the 52- week range. Strong buy from Wall Street, weak buy rating from Seek Alpha. And their semiconductor solutions revenue growth that's accelerated from 11% early last year to 79% in the latest quarter is an extraordinary acceleration for a company already valued at around 1.75 trillion. And total revenue, well that was up 32% year-over-year forecasted to accelerate near 50% forward EBITRA above 50% likewise in fact with EBIT and earnings per share coming in at 59%. Broadcom is comfortably the fastest growing company in today's countdown. But look, it's also exceptionally profitable. Gross margin 76%. EITRA margin 56. Net income margin 39% lever free cash flow margin that's positive 36. Unlike many businesses benefiting from AI investment, Broadcom's already converting revenue into substantial cash and consensus estimates imply EPS could increase dramatically over the next several years. Both these estimates already assume that custom AI chips, networking demand, and VMware integration all progress successfully and revenue estimates will they rise from around 64 billion in 2025 to more than 340 billion by 2030. Now we can see here this demonstrates both the opportunity and the danger. Analysts are not expecting ordinary growth. They're expecting Broadcom to become one of the dominant economic winners of the AI buildout. Now the company also carries around 65 billion of debt against roughly 20 billion of cash and Broadcom's credit default swap spread has recently increased. That signals that some investors are becoming more concerned about leverage and the wider financing requirements of the AI ecosystem and valuation. It does depend heavily on which earnings definition is used. We can see here in fact from simply safe dividends it trades around 23 times normalized forward earnings close to its 5year average and see alpha has it around 32. Regardless the stock's not obviously cheap and if we take a look at the blue tunnel it does sit around the midpoint although look over the last 5 years it spent in fact go over the last 10 all the way since 2023 this one traded dramatically above the fair value. Now, my DCF does contain one particularly aggressive assumption that must be made clear. Although this has been taken from analyst estimates, and that's the free cash flow rises from 27 billion in 25 to 48 billion in 26. If that doesn't occur, the valuation will fall materially. And at the 10% growth rate, we get 320, 13% downside. At 15% growth, 24% upside, 457 value. and at the higher more optimistic 20% it reaches $647 implying 76% upside with the reverse DCF indicating that today's price requires near 12% long-term growth and the middle case does imply near 20% margin of safety. Now Wall Street their average target is $526 representing 43% upside but the range is exceptionally wide from around $216 to $675. Now, this is not consensus certainty. It's evidence of dramatically different assumptions and worth highlighting that Brocom actually reports earnings after the market close on September the 2nd because that event could move the share price substantially. I wouldn't establish a full position beforehand. My approach would be a very small startup position at most followed by reassessment once the results and guidance are available. Broadcom earns third place because its growth and competitive position are exceptional, but its valuation requires an extraordinary AI outcome. And the next company offers less explosive growth, but a far more dependable valuation. And at number two, it is S&P Global. This may be the least exciting name in the countdown, but it could be the cleanest combination of business quality, recurring revenue, and a reasonable purchase price. It owns leading franchises across credit ratings, indices, market intelligence, and energy data. It's still down year to date 15% over the last year down 19%. And it's one we've had on our buyer list since the 52- week low region of $380. Still though, it trades towards that point. 52- week high coming in at $552, more than $100 higher than today's price. Interesting. Our first sale rating from Qu with a disagreement. Wall Street seeded a strong buy, seeing Alpha giving it the weak buy rating. And these businesses operate like financial toll roads. Companies need credit ratings to issue debt. Asset managers license indices to create investment products and financial institutions depend on S&P's data and analytics. Once incorporated into customer workflows, those products are extremely difficult to replace and annual revenues increased around 6 billion in 2017 to more than 16 billion today and annualized growth rate around 11 12%. The company achieved that while building some of the strongest margins in the financial data industry. Let's take a look. Gross margin that sits at 71%. EBITDAR margin exceeding 50%, net income margin above 30% and led free cash flow near 34. These are excellent economics, but the reason the shares have weakened is visible in the Ford estimates. Revenue that's expected around 3.3%. EBIT dollar just shy of 6%. EBIT growth sitting again very similar 5.9 and diluted EPS coming in at 8.8. The market's concerned that near-term momentum is fading and SMB Global also represents around 5% of Bill Aman's disclosed equity portfolio, one that he actually bought in the most recent quarter. Now, it doesn't make the stock automatically attractive, but it supports the view that this is a durable highquality franchise rather than a temporary cyclical trade. And that around 23 and a half times Ford earnings, S&P Global trades well below the 5-year of 29.4, the yield is close to its historical norm. So the attraction here is primarily valuation compression rather than unusually high income. And you can see from the blue tunnels still trades in the undervalued level over the last 5 10 years. First time we really saw this disconnect was pretty much the beginning of the year. Now my base case assumes 8% long-term free cash flow growth and that comes to a value of $531 around 12% above today's price. The low rate of 6% that produces 448 almost exactly at today's price. and the higher rate 10% comes to 629 implying 42% upside. The downside protection is why S&P Global ranks above MSCI and Broadcom. If growth merely slows, the shares already appear close to fair value if the company returns to high singledigit cash flow growth. The upside becomes attractive where we note at the base case a 17% margin of safety and Wall Street's average target $519 implying 17% upside and closely supporting the base valuation. I would consider building a position around $440 while retaining capital in case September volatility pushes the shares closer to 410. So S&B Global is not a statistical bargain. It's an exceptional business experiencing a rare valuation reset and that earns it second place. But the number one company combines an even lower multiple, strong present growth, and the most compelling risk profile in the entire countdown. And my number one stock to buy in September is Meta Platforms. It trades around $578. It's fallen meaningfully from its recent high despite continued to grow revenue, profit, and cash generation. The immediate reason it ranks first is that one of its largest legal uncertainties has now become substantially clearer. And before we get into that, it is down 12% year to date, down 23% over the last 12 months, trading close to 52- week lows with a strong buy from Wall Street, weaker buy from seeking Alpha. And we can see that Med has reached an agreement with a bipartisan group of 52 attorneys generals over youth related social media claims. The agreement imposes new protections, including daily time limits, overnight restrictions, muted school hour notifications, and expanded parental controls. Now the reported financial cost it could reach around 17 billion over 10 years and the numbers enormous in isolation but it represents roughly 1% of Meta's market value and around 8% of a single year's revenue. More importantly it removes the risk of a prolonged trial in which states had argue potential exposure around 200 billion. Now the settlement doesn't eliminate met regulatory risk and product restrictions could affect engagement among young users. However, the cost appear manageable relative to the company's earning power. Operating margin that's around 39% while net margins close to 30. And Meta also retains an A+ profitability grade with gross margins that sit above 80% extraordinary returns on capital and more than 100 billion of cash from operations over the last 12 months. That main weakness though is going to be the aggressive AI investment which has reduced near-term free cash flow growth. Yet revenue is still growing around 28% forward growth although it's slowing still 23% and both EBIT DAR as well as EBIT both very strong numbers earnings per share long-term sitting around 1920 20% at the present valuation Meta doesn't need every one of these figures to continue indefinitely since 2017 revenues expanded dramatically profits increased and the diluted share count has steadily declined therefore combined organic growth with substantial repurchase increasing the owner ship represented by each remaining share and annual profits increased from around 29 billion to more than 120 billion across the period. It makes the current forward valuation particularly notable because look it trades 18 times forward earnings compared with the 5-year 22 that's below S&P Global MCI Brocom and Constellation even though Meta's growth is stronger than most of them and we also get the undervaluation signal on the blue tunnel. Now, let's examine what the market is pricing. My cautious bare case produces a value of $568. That is only 2% below the current price. In other words, if cash flow growth disappoints considerably, Meta appears roughly fairly valued rather than catastrophically overpriced. My base case produces $832, representing around 44% upside. This case assumes Meta's enormous infrastructure investment eventually produces stronger free cash flow through advertising improvements, AI tools, and disciplined expenditure, where the bullish case reaches around $1,179, implying more than 100% upside. Now, I wouldn't use this figure as my expectation, but it demonstrates how powerful the outcome becomes if Meta successfully monetizes its AI investment. But look, the important relationship here isn't the highest number. It's the asymmetry. my cautious case that sits in fact close to today's price. While the base case offers substantial upside, no other company in today's countdown combines that valuation protection with Med's current growth and profitability. And recent analyst target range from around $580 to more than $830 with several major firms remaining positive following the settlement. The broader average though that sits at 755 representing 31% upside is it's below my base case and indicates I'm not relying solely on the most optimistic assumptions. The risk though does remain significant. AI capital expenditure is enormous. Regulatory scrutiny will continue. Realy Labs remains deeply unprofitable and management must demonstrate that infrastructure spending creates acceptable returns. Nevertheless, at around 18 times forward earnings, I believe those risks are more than reflecting the current price. I'd consider building a position purging stages run assuming September volatility is finished. So, Meta earns first place because it offers the strongest combination of present growth, exceptional profitability, valuation support, and upside if management executes. So at number six today we had Next Era that offers defensive exposure to electricity demand and an above average yield but remains highly sensitive to interest rates. At number five we had Constellation which provides more direct exposure to the AI power shortage although free cash flow and valuation require caution. At number four Msei is an outstanding long-term compounder with exceptional margins and recurring revenue. At number three, Broadcom. It offers the highest growth, but also the greatest expectation risk, particularly immediately before earnings. At number two, SMB Global offers arguably the cleanest September entry point for investors, prioritizing business quality and downside protection. And at number one, Meta combines an 18 times Ford multiple with a cautious valuation close to today's price and around 44% upside in my base case. Now, I wouldn't buy all six companies in equal amounts on the first day of September. I'd begin with the businesses offering the strongest downside protection, retain cash for volatility, and reassess Broadcom after the earnings report. September may still produce the volatility history tells us to expect, but volatility is useful when it allows us to buy exceptional business at better prices. So, those were my six best stock opportunities for September, but I want to know which company you would rank number one. Which one you'd avoid completely? Don't forget as well to sign up to the weekly newsletter. We're dropping a fresh copy tomorrow morning. More importantly, have a great day. I'll see you all on the next one.