The Sunday Edge · where the market sits, one stock spotlight, one principle.
The math stock picking cannot fix.
here is a class of investing problems that no amount of stock picking skill can solve. Fees compound. Taxes compound. Turnover compounds. None of them show up in the daily price quote. All three show up in the ending balance 30 years later. The drag is larger than most investors realize, and it runs on the same relentless math that works for you when compounding is on your side.
A 1% annual fee, applied across a 30 year horizon, consumes roughly a quarter of your ending balance. A 2% fee, common in active mutual funds and many advisory relationships, takes nearly half. The drag never feels like a drag in any single year. That is the trap. Structural costs are the rare corner of investing where one decision, made once, keeps paying you for decades. Cutting them requires no forecast at all.
Every reading above is pulled live and set against decades of market history. A single number never tells you what to do, and these are no exception. Their job is context: how the present moment compares with everything the market has already survived.
The returns in this table refresh with the market, so they will look different every time they render. The portfolios behind them do not work that way. They are managed by rule, rebalanced on a fixed schedule, and sized without reference to any single stretch of tape.
MSFTMicrosoft Corporation
No software company in history has carried more distribution leverage than Microsoft. Once a product enters the Microsoft sales motion, the direct enterprise reps, the Azure marketplace, the bundled Office license, the marginal cost of one more customer approaches zero. And every adopted workflow makes leaving harder.
Most software companies fight for distribution. Microsoft owns it. That single distinction explains a remarkable share of why the stock has compounded across two decades of very different regimes.
The moat sits in three layers. First, Office: the productivity stack with no real enterprise alternative, sold as a subscription that embeds itself in daily workflow. Second, Azure: the second largest hyperscale cloud, with a credible position in AI infrastructure that was not obvious five years ago. Third, the AI distribution layer: Copilot, embedded in Office and Windows, monetizing AI compute through the workflows the world already uses.
Every other software vendor has to fight Microsoft's distribution to land a contract. That asymmetry is hard to reproduce, and it shows up in margins that rarely compress, even in down cycles.
The part most investors miss: Microsoft does not have to win the foundation model race to monetize AI. OpenAI models, Anthropic models on Azure, models built in house. The substrate is fungible. The distribution is not. Whichever model wins, the customer who runs it through Copilot pays Microsoft. The market keeps pricing Microsoft as a software incumbent at risk of AI disruption. The company has structured itself as the AI distribution layer for the existing enterprise base. Those are different businesses with very different risk profiles.
Microsoft has been a fixture across our models, carried in both Core 20 and Market Masters. The sizing reflects a simple view: distribution leverage compounds longer than most narratives credit, and most reasonable AI outcomes route through this company regardless of which model wins. The risk worth naming is the bill. Microsoft is spending heavily on AI infrastructure ahead of proven monetization, and regulators keep circling the bundling that makes the distribution machine work. Neither risk is small. Both are the price of the moat. Patience is the trade.
“The price you paid for a stock is information about you, not about the business.”
Anchoring, applied to cost basis
The number on your brokerage statement showing what you paid for a position says nothing about what the position is worth tomorrow. The market does not know your cost basis. It does not care.
Your purchase price is private information about a single decision you made on a single day. It is not a feature of the business.
Yet that number exerts gravity. Below cost: I cannot sell, I would be locking in a loss. Above cost: this is free money, better take some off defensively. Both reactions skip the only question that matters at the moment of any sell decision. Is this business, at today's price, the best home for the capital relative to the alternatives?
The defense is to evaluate every position as if you held cash and were buying it fresh. The cost basis is a distraction. Forward looking thinking beats backward looking accounting. Every time.
Get the next one in your inbox.
The Sunday Edge lands every Sunday morning around 8:30 AM ET. Free. No urgency, no spam. Measured, useful.
Subscribe freeEducational research from Advising Alpha. We are a publisher under Section 202(a)(11)(D) of the Investment Advisers Act of 1940, not a registered investment adviser. Past performance does not guarantee future results. Full disclaimer at advisingalpha.com/disclaimer.