Making Cents Sense

When I look at ACADIA Pharmaceuticals, I see a company that has already crossed one of biotechnology’s most dangerous borders: it has stopped being a clinical-stage promise factory and become a commercial business. That distinction matters. ACADIA has two approved products, growing revenue, positive net income, and enough cash to fund a serious development program without immediately passing a collection plate among shareholders. It also has a pipeline capable of changing the company’s long-term value—assuming, as always, that biology agrees to cooperate. That last condition deserves emphasis. Biotechnology investors have a charming habit of treating every clinical program as an approved blockbuster temporarily inconvenienced by the absence of evidence. A compound enters Phase 2, management displays an enormous patient population, and somebody multiplies that population by the annual treatment cost. By lunchtime, the drug is supposedly worth $12 billion. Actual drug development is less...

For years, I viewed ACADIA Pharmaceuticals as a company with one commercial engine and an uncomfortable amount of hope attached to everything else. That engine was NUPLAZID, the first and only drug approved in the United States for hallucinations and delusions associated with Parkinson’s disease psychosis. It gave ACADIA a real business, not merely a collection of laboratory ambitions. But it also made the investment story painfully concentrated. When one drug carries the company, every prescription trend, reimbursement change, patent dispute, and competitive threat begins to feel like a referendum on the entire enterprise. That is no longer the full story. DAYBUE, approved in 2023 for Rett syndrome, has become a meaningful second franchise. ACADIA now expects more than $1.2 billion in combined annual revenue from two approved medicines, has nearly $1 billion in cash and investments, and is generating quarterly profit while funding a broader neuroscience pipeline. The company has cross...

When I look at Broadcom, I feel as though I am examining two companies wearing the same ticker symbol. One is an aggressive growth machine positioned at the center of the artificial intelligence infrastructure boom. It sells custom AI accelerators, networking technology and other semiconductor products that have become increasingly important as hyperscale customers spend breathtaking sums building data centers. The other is a disciplined cash generator with a long history of raising its dividend, a large infrastructure software operation and enough free cash flow to return billions of dollars to shareholders. Naturally, Wall Street would like me to choose a label. Is Broadcom a growth stock, or is it an income powerhouse? My answer is that Broadcom is a growth stock with an unusually serious dividend habit. It has the financial machinery of an income powerhouse, but its current yield is far too low for me to call it one in the traditional sense. That distinction matters. A retiree seek...

Whenever I study Arista Networks, I run into the same uncomfortable problem: the company keeps producing numbers that make skepticism look increasingly theatrical. Revenue reached $3.036 billion in the second quarter of 2026, up 37.7% from the same quarter a year earlier. GAAP operating margin was 45.4%. Net income was $1.213 billion. Arista ended June with approximately $13.3 billion in cash, cash equivalents, and marketable securities. For the first six months of 2026, operating cash flow approached $2.8 billion. Those are not the financial statements of a business struggling to prove it belongs. They are the numbers of a company executing with the kind of efficiency that makes the average technology executive suddenly remember an urgent appointment elsewhere. Arista has become a central supplier of high-performance Ethernet networking for cloud data centers, artificial-intelligence clusters, campuses, and routing environments. Its Extensible Operating System, or EOS, gives customers...