Should you buy Novartis (NVS) stock in 2026?

2026-08-01 · By Lubin Danilo, founder of Lubin Investment

NVS: see the full analysis on Lubin Investment

Novartis passes 9 of my 10 quality criteria, with a 32% return on invested capital and cash per share that has grown 43% in three years. But the stock trades at 20 times its free cash flow, near its most expensive level in five years, and my model shows a 26.7% overvaluation. Here is my verdict, backed by the numbers.

The largest patent cliff in the group's history

2026 marks, in its own leadership's words, the largest patent cliff in Novartis' entire history. Entresto, its flagship heart-failure treatment, saw its US sales halve as soon as its patent expired, and the cumulative impact of several patent losses this year is estimated at roughly $4 billion in evaporated revenue.

A patent cliff is the central mechanism that governs the entire pharmaceutical industry: as long as a drug is patent-protected, its maker has exclusivity and can sell it at a premium. The moment that patent expires, generic manufacturers can legally produce a near-identical copy at a fraction of the price, and insurers and pharmacies switch to the cheaper option within months. Sales of the original drug then collapse very fast, sometimes 70 to 90% within a year. A pharmaceutical company can therefore never rest on a past success: it must constantly refresh its portfolio of still-patented drugs to offset the ones falling into the public domain.

Is this a good business? (quality)

My filter passes 9 of the 10 quality criteria for Novartis, with a single weak spot: sales growth caps out at 9.3% a year over five years, just under my 10% threshold. The rest of the report card is solid: a 24.7% net margin, a 26.9% free cash flow margin (out of every 100 dollars of sales, almost 27 end up as cash genuinely available after every expense), and above all a 31.9% return on cash invested, well above my 15% threshold. That last figure means every dollar Novartis reinvests in research and plants earns back, on average, close to a third of a dollar of extra cash every year.

The share count has also been falling 3.8% a year for five years, a steady pace of buybacks that mechanically concentrates the value created onto fewer shares. The combined result of all this: cash available per share climbed from about $5.46 in 2022 to $7.80 in 2025, a nearly 43% increase in three years, driven both by the growth in total cash generated (from $12.0 billion to $15.2 billion) and by this steady reduction in share count. This is an accelerating trajectory, not a frozen snapshot: cash-per-share growth even climbs to 18.1% over the last two years alone, above its five-year average.

The bet: can newer drugs offset the older ones?

Second-quarter 2026 results, published in late July, give a first numerical answer to that question. Three newer drugs show spectacular growth: Kisqali (a breast-cancer treatment) jumped 43% in constant currency and topped $1 billion in sales in a single quarter for the first time, with management guiding toward peak annual sales of around $10 billion. Scemblix (chronic myeloid leukemia) nearly doubled, up 89%, driven by 93% growth in the US alone. Leqvio, the first cholesterol treatment based on RNA-interference technology (a molecule that directly blocks the manufacturing of an unwanted protein in the liver, rather than neutralizing that protein once it has already been produced, the way classic statins do), grew 59%.

Novartis reaffirmed its full-year 2026 guidance: low single-digit sales growth for the year despite a first half hit by the patent cliff, driven by an expected acceleration in the second half. This is a bet still being played out, not yet a bet won: the newer drugs are growing fast, but they still start from a smaller base than the $4 billion lost on the older patents.

The price: why does the stock trade near its most expensive level in recent history?

Here is where the story turns counter-intuitive. You might expect a patent-cliff year to scare buyers off and push the stock price down. The opposite happened: Novartis now trades at 20.0 times its free cash flow (P/FCF), a level sitting at the 92nd percentile of its own five-year history, in other words nearly the highest price it has ever fetched over that period. P/FCF, as a reminder, is simply the stock price divided by the cash the company generates every year: the higher the number, the more the market is paying for every dollar of cash generated.

It is not that cash generation has declined: I just showed it has actually grown nicely. It is the stock price that has climbed even faster than that cash. The market is not betting on decline, it is betting the opposite: that Kisqali, Scemblix, and Leqvio will fully offset and then exceed what is being lost on Entresto and the other expired patents, and it already accepts paying top price for that confidence, even before the second half of 2026 confirms or disproves that trajectory.

My model, which projects the cash-per-share trajectory over five years using conservative assumptions rather than extrapolating recent growth indefinitely, sets a fair buy price of $114.45. The current price of $156.15 therefore represents a 26.7% overvaluation relative to that price. This is not a verdict of distrust in the business: the quality of the story is not in question. It is a disagreement over the pace at which today's market confidence should be paid for, before the evidence from the next two or three quarters is fully in.

A dividend that rounds out the picture

Novartis pays a dividend yielding 3% at the current price, with a payout ratio of 71.6% of earnings, growing 7.2% a year on average over five years. That is a reasonable payout pace for a mature pharmaceutical company: neither a red flag (a payout ratio near or above 100% would signal a company distributing more than it earns), nor a token dividend that would not genuinely reward shareholders. This dividend adds to the buybacks already discussed as a second channel of shareholder returns, both funded by available cash that keeps growing despite the transition year.

My method, in short

I always keep two questions separate that most investors blend together: is this a good business, and is now the right price? For Novartis, the first answer is clear: 9 out of 10 criteria, an excellent return on capital, cash per share that is accelerating. The second answer is more nuanced: the price already bakes in much of the optimistic scenario on new drugs, before that scenario is fully confirmed in the accounts. I would rather wait for the second half to confirm the trajectory, or for a pullback to bring the price closer to my estimate, than pay today's market confidence at full price. You can track these figures live on Novartis' page, and see exactly how I calculate this fair buy price in my methodology guide.

Novartis passes 9 quality criteria out of 10: a 31.9% return on invested capital, a 26.9% free cash flow margin, cash per share up 43% since 2022 ($5.46 to $7.80).
2026 is the largest patent cliff in the group's history: Entresto's US sales halved, for an estimated cumulative impact of $4 billion this year.
Three growth drivers are advancing fast to offset it: Kisqali (+43%, over $1 billion in a single quarter), Scemblix (+89%), Leqvio (+59%).
The stock trades at 20.0 times its free cash flow, at the 92nd percentile of its own five-year history (nearly its most expensive), with the market already paying for confidence in this growth relay.
My model shows a 26.7% overvaluation (fair buy price of $114.45 versus a $156.15 price): the quality is not in doubt, the price paid today for the market's confidence is.

FAQ

What is a patent cliff in the pharmaceutical industry?

It is the moment a drug's patent protection expires, letting generic makers sell a copy at a much lower price. Sales of the original drug then collapse very fast, often 70 to 90% within a year, as insurers and pharmacies switch to the cheaper option.

Why is Novartis stock expensive when 2026 is a tough patent year?

Because the market is not looking at the past but at the future: it is betting that fast-growing newer drugs (Kisqali, Scemblix, Leqvio) will offset and then exceed the patent losses, and it already accepts paying top price for that confidence before the evidence is fully in.

Is Novartis undervalued or overvalued in 2026?

According to my model, it is overvalued by about 26.7%: my fair buy price sits at $114.45 versus a $156.15 price. The company's quality is not in question, it is the price demanded today for confidence in the new drugs that looks rich to me.

Is Novartis' dividend solid?

Yes: a 3% yield with a 71.6% payout ratio of earnings, growing 7.2% a year on average over five years. That is a reasonable level that leaves room, neither a red flag nor a purely token dividend.

Should you buy Novartis stock in 2026?

The business is very good quality (9 out of 10 criteria), but my model shows the current price already bakes in much of the optimistic scenario on new drugs before it is fully confirmed in the accounts. This is not personalized investment advice, do your own research.

About the author

Written by Lubin Danilo, founder of Lubin Investment. A self-taught individual investor, I find fundamental analysis fascinating, and it has delivered excellent results. For three years now, my performance has beaten the S&P 500. But analyzing every stock took too much time: sites with incomplete data, calculation methods and criteria never aligned with mine. And spotting the best stocks was just as time-consuming, even with my own well-defined checklist. So I put my software development background to work to build this software, base my investment strategy on its results, and share it with people who share the same passion as me. It judges a company's quality and its price separately, using criteria drawn from the financial literature (Warren Buffett, Michael Mauboussin, Aswath Damodaran).